Linux 2.6.5 is Released

lamont116 writes "Featuring a 367.6 KB changelog, the next Linux kernel is now ready for action. As always, enjoy!"

Linux Changelog Email Publishing

[dolo666]

With all the Spam problems going on these days, you would think that the Linux changelog would not publish email addys. Why should someone have to display their email addy to work on Linux? To me it seems very counter-productive, and it may shed light as to why Linux users wish to stop Spam, instead of simply becoming inaccessible to it. I've thought about working on Linux before, but this is the only thing stopping me as it stands.

the PPC64 work is looking real nice

[phoxix]

40+ patches, ouch :)

One day PPC64 will be just as common as AMD64 in the server room ...

Sunny Dubey

Comment removed

[account_deleted]

Comment removed based on user account deletion

Instructions for 2.4 to 2.6 upgrades for Luddites?

[JessLeah]

Luddites like me might like to try 2.6, if only we had some guidance. I still run Debian Stable (yes, Stable) since I don't trust Unstable or Testing not to muck up my system.

Anyone got upgrade instructions for Debian 3.0, or other 'old' distros? Believe it or not, not everyone wants to be on the 'bleeding edge' in all areas. Nevertheless, to be able to try new kernels would be nice.

Where can i browse online through the source?

[Anonymous Coward]

Probably a stupid question as everyone except me knows where this is possible, well nevertheless...

is there a site where i can just browse through the sources without downloading them?

Re:Linux Changelog Email Publishing

[dolo666]

> Go on then, scratch your own itch.

Okay! I'll write something to handle this, and post the source code at my site next week when I can free up some time. Maybe I'll write a changelog writing app and put in some extra features so the Linux team uses it (and anyone else)...

Installation?

[black mariah]

Does anyone have links to some articles that details kernel upgrades that aren't written by and for programmers? I really want to update my Fedora box to 2.6, but the documentation I've seen for installing a new kernel pretty much assumes it's something you've done before.

DMA API changes?

[Mithrandir_The_Wise]

According to this article [lwn.net] on LWN.net there was a patch by Dave Miller that changed the DMA API (see the Changelog for 2.6.5, from Dave Miller submitted through Andrew Morton) which *might* break binary drivers. All the in-kernel drivers are fixed, but the out-of-tree stuff might screw up. Just a heads up.

Scheduler?

[moxruby]

After reading the hype regarding the new kernel, I installed mandrake 10.0 (k2.6.3 I believe) to check it out. I was disapointed; mp3s skip under light loads even after I raised the priority of the player to maximum. I've got a 1.3ghz duron w/ 256mb ram so the machine should be able to cope.
I googled for a mandrake-specific bug but found nothing... Anyone else had the same problem?

version.h

[jmoen]

Seems like version.h is missing from include/linux in this release (I patched from 2.6.4).
Made my ATI binary driver fail the compile, though by copying my 2.6.3 version.h file fixed it.
I thought the version.h file where used by several other prgograms that compiles against the kernel.
Am I missing something obvious here ?

Re:Cut and Paste

[roror]

I doubt anyone understands that generic joke. Aside, it probably the first distro which will "support" the kernel.

Reiserfs

[EvilTwinSkippy]

Have they fixed the goofiness with reiserfs and 2.6. On my RAID array I get random oddities, like a user who owns a file but can't access it. And our Veritas Backup Exec client won't decend down to a reiserfs mounted volume.

Sure I could wipe and reformat the array, but the bugger is 100GB and my last restore took 48 hours.

Alan Smithee is going beyond films ...

[anti-NAT]

The name Alan Smithee [imdb.com] is commonly used by film directors who's control of the film was taken away from them, such that the director wants to disown it.

Re:Linux Changelog Email Publishing

[Anonymous Coward]

In this case, I agree, the mods did a wrong, but I've done similar modding in the past. Sometimes it's nessicary due to the mods being partial to one side of an argument. Example:

Person 1: (insightful question/comment against outsourcing) +1 Insightful
Person 2: (Obvious trolling in favor of outsourcing) +5 Interesting
Person 1: (Valid counter point to person 2) +0 Offtopic
Person 2: (Counter-point to person 1 laced with anti-American flambait) +5 Insightful

In such a case, I mod the higher-modded person as overrated, and the lower-modded person as underrated.

Always drivers...

[r6144]

Well, I'm running 2.6.x on two of my machines now, and they are running mostly perfectly (user-mode-linux doesn't work well for me yet, as of 2.6.3). Anyway I did have a (very old) machine in which 2.4 kernels fails to detect the network card correctly even after tons of isapnp tweakings, so I had to downgrade the kernel to 2.2 after upgrading RH7.0 to RH7.3.

Such things depends mostly on luck, since obviously it is the drivers that are problematic, and some hardware are owned by few kernel hackers, so hard-to-fix kinds of bugs in them can take much time to fix, while it is reasonable of Linus et al to start flagging the kernel as "stable" if it works on 50~75% of the machines.

It seems that there are more hardware companies than excellent kernel hackers for many operating systems (maybe even Windows), so driver quality will always be a problem on any OS for a long time to go...

Re:Installation?

[Telex4]

You might find this document useful:

http://www.newtolinux.org.uk/tutorials/linuxkern el .shtml

It introduces all the basics of the kernel, including what it contains, and how to start playing around with modules. At the bottom there are some links to tutorials on compiling your kernel, and then setting up your bootloader.

In all, they should get you through all except odd problems.

VM/swapd

[evac]

From the "I-might-have-to-run-this-in-production" department, VM patches like this in a stable tree give me the roaring shiznits:

[PATCH] Narrow blk_congestion_wait races
[PATCH] kswapd throttling fixes - comment "The logic in balance_pgdat() is all bollixed up"
[PATCH] shrink_slab: math precision fix - comment "In shrink_slab(), do the multiply before the divide to avoid losing precision." !!!!
[PATCH] vmscan: avoid bogus throttling
[PATCH] fix the kswapd zone scanning algorithm ...etc etc. Many more of the same.

If we're weren't currently having VM issues with 2.4 (servers with 8gig+ memory) I wouldn't care. But we're seriously thinking of using 2.6 in production to resolve it. No, stop laughing, really.

Actually I don't think I'm going to read any more kernel changelogs. It's like being at a restaurant, sometimes you just don't want to know what's going on in the kitchen. Except with open source, the kitchen is more like a public urinal. And the food is one big shit sandwich that everone... ok I'll stop now.

Re:Performance...

[Featureless]

If you have sarge or sid, why wait? It's too easy!

I was able to apt-get the 2.6.3 release, and it "just worked."

Basically you can blast away at the disk and the system feels as solid as a rock - barely hesitates at anything.

It made windows seem so annoying by comparison that I finally switched.

Again

[Apreche]

I must say again the same thing I said the last time the kernel updated. Make a USEFUL CHANGELOG. That huge document is completely useless to a guy like me who needs to decide whether or not to rebuild his kernel or not. Make a changelog more like winamps that lists what actually happened. For example

* X piece of hardware now works
* We made X faster
* X is more secure now
* X is more stable now

stuff like that is useful because I can grep for X and see if anything I use is changed. With the current changelogs I get stuff like.

I chose this at random

<[e-mail]>
[PATCH] netconsole init return code

tiny-mpm/drivers/net/netconsole.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)

What the hell does this mean? Obviously the netconsole return code was changed. 2 lines were added and two lines were removed. But wtf does this mean for the end user!!?! You know the guy who compiles and uses the kernel, but never hacks the source. Make a changelog for me!

Re:Linux Changelog Email Publishing

[Anonymous Coward]

You like that word "systemic," don't you? Trolling dumbfuck.

The act of shrugging off spam is NOT what made it prevalent today.

online publication of email addresses encourages spam

This is a complete bunch of crap. Publishing an email address does not support spammers. Spammers only continue because they get a return on their advertisements. So spam is directly a result of the consumers of spam advertised goods, NOTHING ELSE. Furthermore, the people buying spam sold products are going to be giving up their email addresses anyway. A bunch of kernel developers and others that are using email for what it is was originally intended do not represent even a millionth of the email addresses out there.

You are a fucking idiot. Living as some new age buddhist babbler doesn't change that.

DMA

[Bob_Robertson]

hdparm -d1 /dev/hda

This has bitten me enough times in the past that I assume it any time a "media" file skips. Thanks for posting it first, from the rest of the comments it didn't occur to too many folks.

Bob-

Re:the PPC64 work is looking real nice

[niko9]

Anybody know why? Does he --or anybody at IBM-- see a future for PPC64 and Linux as a desktop contender?

This [theinquirer.net] is why I ask.

And with IBM's recent "open" hardware initiative for PowerPC, things are looking tantalizing: Open OS and Open hardware.