Linux 3.0

An anonymous reader writes "In a post to the kernel mailing list, Rob Landley, sitting in for the floating Linus, cracks the whip over what will be in Linux 3.0. His orders are on Linux and main."

Comments

The floating Linus?

[mccalli]

...sitting in for the floating Linus...

He's achieved a transcendental state now? What are the kernel people going to do when he finally ascends to Nirvana?

Cheers,
Ian

Re:The floating Linus?

[SkulkCU]

For those wondering, he's on a boat.

Either way, he doesn't exactly have his feet planted firmly on the ground...

Damn.

[packeteer]

And 2.4.19 is STILL compiling on my 50 mhz box...

Re:Damn.

[packeteer]

How was that a troll? Honestly it takes a long time to compile linux on an old box. Im not saying thats bad. You ever compiled windows? Ive seen it done and it WAY mroe painful.

Re:Damn.

[stud9920]

Just compile the much shorter Linux 0.1.81 (this will take much less time), diff it with an empty file giving you Linux -0.1.81 and diff that one with Linux 2.4.19 giving you Linux 2.4.19 -(-0.1.81)=Linux 2.4.19+0.1.81= Linux 2.6.0. If they choose to call it Linux 3.0, do the same trick with Linux 0.5.81 instead.

Ah,those newbies. Next time RTFM !

Re:Damn.

[Fulcrum of Evil]

You ever compiled windows?

Yeah. It only took a couple of hours on an 8-way dell box that cost my yearly salary. To be fair, this was a few years back, right after I graduated.

3.0?

[user32.ExitWindowsEx]

You mean it was determined the kernel is going to be called 3.0 instead of being called 2.6 after all?

Re:3.0?

[InodoroPereyra]

You mean it was determined the kernel is going to be called 3.0 instead of being called 2.6 after all?

No, the article says "3.0-pre (or 2.6-pre) series". And what's the big deal after all ?. Call it 2.6, 3.0, whatever :-)

Re:3.0?

[4of12]

Of course there's long been a convention in Linux land that less stable development kernels have odd numbers like 2.1, 2.3 and 2.5, while even numbers denote the stable series meant for pedestrian users. [Although many could argue that the VM switch during 2.4 did not exactly belong to a stable series.]

Anyway, if we're going to have an odd number major version, then all I can say is

"Get ready for a ride"...

Re:3.0?

[Kevin Stevens]

Well, considering all the work that has been put into all the "minor" version upgrades, especially 2.2->2.4, I think there is an expectation that any major kernel version jump should have a considerable number of new major features included in it. From a geek POV, its not that big of a deal, but to some boss type making decisions on something he has only a vague grasp of (bigger version numbers are better than smaller ones) It is important, and if somewhere he reads that the change from 2.whatever he is using to 3.0 doesnt include anything of signifigance, he/she will then be dissapointed in Linux's development. If you want to encourage widespread adoption, you have to play by the business world's rules. It is unfortunate, yes, but what can you do? And if Linux is reduced to a small niche community supported OS, progress will slow down quite a bit, especially in areas like enterprise computing where companies like IBM have been throwing a great deal of weight behind the OS.

Re:3.0?

[Anonymous Coward]

What Linus means is that the sum of the major and minor version numbers denote stability. Version 1.3 adds up to 4 which means that it is a stable branch, 2.1 is unstable because it adds to 3. In that case, 3.0 would be the first developmental release (probably a symlink to 2.5), while 3.1 is where the real action begins. Note that 3.1 happens to be the universally accepted branding for an unstable release as evidenced by another "operating system" durning that stage of development.

Re:3.0?

[SurfsUp]

You mean it was determined the kernel is going to be called 3.0 instead of being called 2.6 after all?

It was not. Linus will decide that, not Rob Landry.

Re:3.0?

[iabervon]

Ah, but this time, the release management isn't going to be done by Linus, but by (IIRC) Marcello. In order to make this fact clear to people who don't follow the development process, he's going to skip right from 3.0-rc(NN) to 3.0.12, at which point it will be stable and go into mostly maintenence mode, with 3.1 starting up.

Re:3.0?

[mfarver]

I've met Rob... even had the misfortune of trying to figure out a major project coded by him. Basically he is a nice guy, and a fairly smart one, but is also a bit of a blowhard, and prone to inflating his own importance to stratospheric levels. I would take anything written by him with a grain of salt.

Re:3.0?

[landley]

Huh. I'd have a better idea which project you were talking about if I had the slightest clue who you were. :)

Rob

3.0? bah

[exspecto]

i can't wait till Linux 3.11 for workgroups

Re:3.0? bah

[Zephaniah]

Why is this modded up as funny? I swear I've seen this gag every single time there's an article about any software that's at version 3. Gets a bit old is all I'm saying.

Re:3.0? bah

[FrankNFurter]

DOS-Based Windows versions:

Windows 95 = Windows 4.0.950
Windows 95B = Windows 4.0.1111
Windows 98 = Windows 4.1.1998
Windows 98 SE = Windows 4.1.2222
Windows ME = Windows 4.9.3000

NT-Based Windows versions:

Windows 2000 = Windows 5.0.2195
Windows XP = Windows 5.1.2600

HTH.

Re:3.0? bah

[Amazing Quantum Man]

And of those, the only decent ones are

98SE and 2000 (4.1.2222 and 5.0.2195).

Re:3.0? bah

[Dave_bsr]

Look! I have an opinion! It's at least somewhat right! It's just a little bit negative...I must be a troll! whee!!!

Nazi mod alarm: ding-ding-ding!!!

Actually I agree with you. 95 is best left to die. 98 is decent. 2K really is the best so far. XP...well..we all know about XP.

Re:3.0? bah

[drinkypoo]

Incidentally, 95B was OSR2; There IS an OSR1 which is fairly insigificant, it has some fixes on top of the original OEM release. There was also an OSR3 beta which became Windows 98 beta. Unfortunately I don't know its build number, it was a 25 day beta and I ran it for 25 days and went back to OSR2 (At the time) and then to a Win98 RC later. (*whistles casually*)

Re:3.0? bah

[Shanep]

And of those, the only decent ones are

98SE and 2000 (4.1.2222 and 5.0.2195).

Actually I agree with you. 95 is best left to die. 98 is decent.

What the hell are you guys smoking?! : )

Win2k is great, and NT4 OK, comparitively speaking. But ALL of the DOS based Windows suck severely. I've been supporting them all since 3.0, I'm amazed there is *anyone* who thinks 98 is at all decent.

XP is just incredible. Decent foundation with 95 stability and quirks. I fail to see how MS could possibly have made it so damn bad.

My most anticipated feature

[Sheetrock]

LKCD: Linux Kernel Crash Dumps. Really, I wish this had been there for the first half of 2.4 (testing-pre?). Supposedly it'll be able to save an image of kernel memory when the kernel panics to a special partition so that it can be recovered after reboot allowing easy analysis of the image. This alone should cut down greatly on the amount of work required to submit bug reports.

Re:My most anticipated feature

[NecrosisLabs]

What would be great would be if it automatically piped it to the screen, with some form of high contrast text, like white text against a blue background...

Re:My most anticipated feature

[pcidevel]

I develop drivers for Solaris, HP-UX, and Linux (for a PCI card that my company makes). I would love to see a Kernel dump for a particular problem I am trying to debug at the moment! :)

Of course, stupid mistakes in Solaris or HP-UX kills their kernel and results in waiting for the machine to reboot. Stupid mistakes in Linux results in a kernel panic with the output sent to the syslog (9 times out of 10 bad code doesn't kill the entire Kernel, so no waiting on the machine to restart), so I definately think that Linux has the upper hand as far as handling poor kernel space code.

Re:My most anticipated feature

[dracken]

Bah! all you geeks are struggling to make the first gen BSOD. Look at Cowboyneal's article last friday

Posted by Cowboyneal on Sunday October 20, @11:16AM
from the yippee-new-windows-features! dept.

In a surprise announcement today, Microsoft President Steve Ballmer revealed that the Redmond based company will allow computer resellers and end-users to customize the appearance of the Blue Screen of Death (abbreviated BSOD), the screen that displays when the Windows operating system crashes.

The move comes as the result of numerous focus groups and customer surveys done by Microsoft. Thousands of Microsoft customers were asked, "What do you spend the most time doing on your computer?" A surprising number of respondents said, "Staring at a Blue Screen of Death". At 54 percent, it was the top answer, beating the second place answer "Downloading Pornography" by an easy 12 points.

"We immediately recognized this as a great opportunity for ourselves, our channel partners, and especially our customers." explained the excited Ballmer to a room full of reporters.

Immense video displays were used to show images of the new customizable BSOD screen side-by-side with the older static version. Users can select from a collection of "BSOD Themes", allowing them to instead have a Mauve Screen of Death or even a Paisley Screen of Death. Graphics and multimedia content can now be incorporated into the screen, making the BSOD the perfect conduit for delivering product information and entertainment to Windows users.

The Blue Screen of Death is by far the most recognized feature of the Windows (tm) operating system, and as a result, Microsoft has historically insisted on total control over its look-and-feel. This recent departure from that policy reflects Microsoft's recognition of the Windows desktop itself as the "ultimate information portal." By default, the new BSOD will be configured to show a random selection of Microsoft product information whenever the system crashes. Microsoft channel partners can negotiate with Microsoft for the right to customize the BSOD on systems they ship.

Major computer resellers such as Compaq, Gateway, and Dell are already lining up for premier placement on the new and improved BSOD.

Ballmer concluded by getting a dig in against the Open Source community. "This just goes to show that Microsoft continues to innovate at a much faster pace than open source. I have yet to see any evidence that Linux even has a BSOD, let alone a customizable one."

Re:My most anticipated feature

[GiMP]

Not to mention how long it takes for some of those 'big iron' machiens to boot ;) My HP9000 takes a good 10-15 minutes to reboot.

Re:My most anticipated feature

[cpeterso]

As a device driver developer, do you think Linux's interal APIs change too frequently compared to other Unix operating systems? Would there be much value in creating a more stable, possibly cross-platform driver API for Linux? How much cross-platform driver code can be shared across Solaris, HP-UX, and Linux?

Re:My most anticipated feature

[Pathwalker]

15 minutes for a HP9000? That's nothing.

I used to work with a cluster of V class systems where each node would take over 90 minutes to boot.

Just bringing the cluster back up if it was completely shut down would take about 15 minutes from the cmruncl until all packages were up and running.

Re:My most anticipated feature

[pcidevel]

As a device driver developer, do you think Linux's interal APIs change too frequently compared to other Unix operating systems? Would there be much value in creating a more stable, possibly cross-platform driver API for Linux? How much cross-platform driver code can be shared across Solaris, HP-UX, and Linux?

Well, I've only supported 2.2 and 2.4, and there were quite a few changes between them, but, for the most part, you can write a 2.2 compatible driver and it will run on a 2.4 kernel. Luckily my company didn't see any benefit to supporting the 2.2 kernel after 2.4 came out (we sell a very costly, like $10k, PCI card that performs a very specialized function, so we work hand in hand with our customers, which means we don't have to have much backwards compatability most of the time). If I were developing a card that was sold to consumers, and I needed to be able to support a wide variety of machines, then I would be much more likely to answer yes to the stable API question! :)

As far as sharing cross platform code, well we originally used streams on Solaris and HP-UX, but we found that the streams support was lacking (at best) on Linux, so we lost a lot of time and code when we originally ported to Linux. Now we basically have 2 sets of Unix drivers. A streams version for Solaris and HP-UX, and a non-streams for Linux. We're hoping to remove all of the streams code and re-merge the drivers at some point, but it's always hard to balance a task like that vs our other time commitments. The differences between our HP-UX and Solaris drivers are a few #ifdef's, whereas the difference between that driver and Linux is probably thousands of lines of code.

Re:My most anticipated feature

[einhverfr]

Agreed, perhaps because most of the drivers are really high quality. BUT for development, it is extremely handy to have these esp. in the developer and -pre releases.

I will be very uncharacteristic and say that the only kernel panic I saw was from installing NT on a Linux box without removing LILO.... Same reason, different cause. Needless to say, I installed Linux again lot long after ;)

WhooHoo

[papasui]

Ready - Rewrite of the console layer (James Simmons) http://linuxconsole.sourceforge.net/
This one specifically should significantly help Linux take off on more devices.

Linux 3.0

[TonyZahn]

I think that, in response to current marketing trends, Linux 3.0 should be given a 2-letter version id instead of a number.

How about:
Linux IS (For those unbelievers...)
Linux ME (friendlier, bloatier, used like a verb)
Linux XL (for those kernel with everything)
or
Linux ** (just take care of all the letter names at once)

Why do i care?

[nurb432]

Not to sound like a troll, or flaming developers, but seriously, from a users standpoint, why do i care?

What i have now works great, give me concrete reasons i should worry about a new release.

Now as a developer i DO care.. I'm just looking at this from the stand point of a normal user ( my customers ) who hear the same stuff from M$ or apple.. 'new and improved, you must upgrade now'... And we used that as a selling point for Linux..

Re:Why do i care?

[juju2112]

As a developer, you are a user of the linux kernel. Most people who actually use linux and know things about it are going to care. Cherishing the opinions of people who know nothing about linux at all is a very odd way of thinking about it, if you ask me.

Perhaps Steven Spielberg should not care about the latest developments in movie cameras beccause I can't tell the difference?

Re:Why do i care?

[FooBarWidget]

Because Slashdot is a site for technical people. They care. I'm not a kernel developer, but I care about Linux development.
If you don't care, then:
1) You do not belong here. Go find yourself a different news source.
2) Change your account settings to hide Linux-related stories.

Re:Why do i care?

[Chanc_Gorkon]

Absolutly. If what you have now works, then by all means, don't upgrade......with a caveat. As with any OS, after 3.0 or 2.6 whatever it's going to be called comes out, then bug fixes may not be done for that much longer to the 2.4 series. After a while, you will probably want to upgrade anyway especially if your company pays for support from Red Hat or whoever.

The upside with Linux is that with every new version, you usually don't have to upgrade hardware all that much or at least as much as say Windows.

My only complaint.....some installers now (Red Hat's and Mandrake's in particular) won't let you install on a low end machine ( I know there may be other versions, but I am talking about the default installer.....). By low end, I mean 486 and Pentium (No Bloody II, !!! or 4). Granted, this don't hurt many, because those in the know can just get Debian and install it.....but what makes Red Hat and Mandrake so certain that you can't get something to run on those machines? It seems, to me, that maybe if they had one low end image that let you just install it anyway and just deal with the circumstances afterwards would be a better thing to do. Like I said, for most, it doesn't matter. You can pick up Pentium II's (old machines) for peanuts now, so that guy can go and upgrade that decrepit Pentium 100. But my point is, why be like Microsoft and force ANY upgrade? At some point, you could drop that support, but there are alot of those machines laying around yet and they can serve a purpose before going to the landfill.

Re:Why do i care?

[dattaway]

why do i care?

The kernel is the framework that connects to the hardware. Its like the chassis on your car, designed to give all components a secure mounting point to connect to each other. Without the kernel, all your component software would still be functional, but not to useful as they would be laying on the ground in a proof of concept state. The kernel supplies all the hardware to hook things up and make them into a fully functional machine.

The init process and scripts, libraries, and applications are the engine, powertrain components, interior, and all the other details to make a complete operating system. The kernel is simply the framework and body to make it all possible. Compile options allow you to have lightweight race car or a dumptruck.

Re:Why do i care?

[sg_oneill]

No Dude. That's unfair.
If as you claim , you are a developer, then you know full bloody well the importance of keeping up to date with operating systems.
And as for "go away". It's only as rude as "I don't care"(Which is code for "shutup" when you think about it).
And Duh... This *is* a linux orientated site. That's why linux news is important.
And when your users systems get ripped to shit by hackers because Redhat 5.2 was still "working for them", what do you usually explain to them?

Re:Why do i care?

[Wdomburg]

>Absolutly. If what you have now works, then by >all means, don't upgrade......with a caveat. As
>with any OS, after 3.0 or 2.6 whatever it's going
>to be called comes out, then bug fixes may not be
>done for that much longer to the 2.4 series.

You'd have to worry more about new features and hardware support getting in than bug fixes. The 2.0 kernel is still being maintained and its over six years old.

>After a while, you will probably want to upgrade
>anyway especially if your company pays for >support from Red Hat or whoever.

If you're paying for support from a company, you'd probably be best off running whatever kernel that's current for the distribution, unless you have some specific reason to deviate. And with the lengthened life cycle of products like Advanced Server, the likelyhood of having to move to a new kernel for support reasons is even lower.

Matt

Re:Why do i care?

[FooBarWidget]

And why should I care about what the masses think? This is SLASHDOT, 99.9% of the masses don't come here! This is a site for nerds, so I give my opinion as a nerd. If I have a job, then I adapt my attitude to the situation. If you're in Rome, do as the Romans.

Think whatever you want about the Linux community. I told you I'm representing the Slashdot community, and you realize that the small impression may not be always right, yet you still hop behind the masses. Popularism isn't going to get you anywhere. Letting the masses do all the thinkwork for you is stupid. Wether that's the reality or not is irrelevant.

Re:Why do i care?

[Chanc_Gorkon]

This is the problem. A 486 DX4100 is not THAT much slower then a pentium 100. Sure, it's floating point sucks much worse then a Pentium, but there's the rub. In any case, Debian is the only one I know of that will install on lower then pentium hardware. Sure, your not going to be playing games on these things. A faster 486 machine could easily perform as a firewall. There maybe some firewall only stuff that works just fine, but I am not aware of those.

Re:Why do i care?

[drinkypoo]

People who are paying for linux are not paying for the broadest hardware support. They're paying for an easy experience and the specific features they want. These features do not include running on ancient hardware.

People who bother to blow the dust off the laptop they got at a yard sale are able to run, say, gentoo on it, which is especially handy if it's a K6.

Re:Why do i care?

[FooBarWidget]

No, and stare at a screen with only stories about Science, Patents, Censorship, Space, Microsoft, etc.

Re:Already got a beta version....

[cnkeller]

Here is a good place to find out about the state of various features...

roadmap?

[Khopesh]

so 2.6 will be 3.0 ... i hope there is effort to make it stabler and more efficient than the 2.5 builds i've tried.

kernel.org doesn't have any info on this...

does anybody know of a roadmap (iirc, there is no official one)
or good guess as to when 3.0 (or 2.6) will be released?

Re:roadmap?

[javahacker]

The 2.5 builds were development builds, they are supposed to be unstable, because they are full of code that is being tested.

Everyone, especially Linus, hopes that the 2.6 (or 3.0) series has a more stable start than the 2.4 series did.

The real answer is when Linus and the core kernel developers believe it is ready, and there is no schedule, because they don't know when that will be. The answer I believe everyone is hoping for, is by the end of the year, since that will let it be included in the next release cycle by the major distributions (Red Hat, SuSE, Mandrake), who are probably targeting releases late in the first quarter of the year.

Let me be the first to come out with the bad joke

[GreyWolf3000]

Again, some of the things on this list won't make it into 3.0. It's just candidates. But everything that is NOT on this list in about 7 days is probably going to become 3.1 material by default.

Speak now, or till 3.1 hold your peace...

Ok, lets all acknowledge the obvious cracks at 3.11 (like what happened with Windows). Let's sort of communally agree that we're not going to find 'em funny, before a really dumb thread enters the picture, okay?

Not Version Bloat.

[muixA]

Before anyone gets going on it.
There have been MAJOR features added to this Kernel.
Including
- UML
- New VM
- New Scheduler
- Finer SMP Locking
- At least 2 new Journaled FS (Reiserfs4 and XFS?)
- A new POSIX thread library/API.

Does anyone know if ALSA will be included?

We will finally be able to forget about the 1980's style /dev/dsp :)

--
Matt

Re:Not Version Bloat.

[bripeace]

Alsa is already in..

Resier4 isn't in yet.. but he plans on submitting it on the 27th.. at least thats what his last email said.

XFS is already in there though

Re:Not Version Bloat.

[macdaddy]

We will finally be able to forget about the 1980's style...

But I liked the 80's style. I haven't worn out my baggy denim jacket yet and my spike hanging in there as good as ever. On a sidenote, to you men with big bald spots, I've been told that a spike is one way to conceal a growing bald spot. Become a PHB now! I want my IIe back.

Re:Not Version Bloat.

[iabervon]

Not to mention all of the other features that have been added between 2.0 and 2.4, all of which should contribute to the decision to change the major number. It would actually be really interesting to see how much of the code hasn't been changed during the 2.x series.

Re:Not Version Bloat.

[rweir]

ALSA is defintely in, as is things like USB2, Access Control Lists, new NTFS support (it doesn't completely trash partitions now!), hot pluggable CPUs, software suspend (hit a key combo and save the whole system state to the harddrive), support for drives >2TB, and a whole lot more.

BTW, I have to love a community where this sort of thing is discussed on a site called KernelNewbies:)

Re:Not Version Bloat.

[be-fan]

support for drives >2TB
>>>>>>>>
Dammit Maxtor, I just bought the 160GB model the other day!

Re:Not Version Bloat.

[Fnord]

Actually does anyone know whether this version will have a real AIO implementation? The current version of AIO is implemented completely in userspace (in libc, emulated ontop of pthreads). Will we finally have real asynchronous IO?

Re:Not Version Bloat.

[LunaticLeo]

Yes it is in. Ben LaHaise is doing this. See linux-aio list, or ben's web page at redhat (google it; I don't recall it right now).

cruise

[Deton8]

Screw the kernel, I want to hear more about these geek cruises!

Re:cruise

[macdaddy]

Not if the female wildlife onboard is geek women. What we need is a geek cruise with non-geek women . :-)

Re:cruise

[dattaway]

I don't know about geek women, but this appears to be sponsored by the same people who threw extravagant and very memorable parties at the ALS in Atlanta over the years (Linux Journal.) I suspect many of the women I saw were in sales at the many booths and oh did they look good. Did you know kernel hackers could put on a show and dance too? You might be surprised.

Is media automount in the kernel yet?

[Bonker]

Automount of removeable media like every other modern OS - Windows does it. MacOS does it. Even DOS 6.2 did it. Why doesn't Linux automount (please note that I did not say 'Autoplay') removeable media? (Note, I only use 2.4 kernels in servers. This may have changed recently, and I justed missed it, but...)

Re:Is media automount in the kernel yet?

[BJH]

Check out something like WOLK - it includes the Supermount patches, which do exactly what you want (as well as automount of NFS shares upon first access).

Re:Is media automount in the kernel yet?

[Rastor]

Why doesn't Linux automount (please note that I did not say 'Autoplay') removeable media? (Note, I only use 2.4 kernels in servers. This may have changed recently, and I justed missed it, but...)

In my Red Hat servers I do this with the autofs daemon. I've used it successfully with both CD-ROM and ZIP drives, and had no problems sharing automounted drives with Samba.

Mandrake has for years now (with the exception of the 8.1 release, I believe) gone one step further by using a kernel patched with "Supermount", which is a "true" automount like you are probably thinking of.

So in fact, Linux has had this ability for quite some time now.

Re:Is media automount in the kernel yet?

[Ektanoor]

Are you from Mars? There are two versions of automount in Linux and there is something called "supermount". But if you use Linux for servers, then you are deeply wrong if you put these things into action. The automount feature is not embedded to Linux just for that reason. A good secured server should in no way give a chance to automount third party media. Only the admin should do it and he shall have a chance to do it flexibly and correctly. Believe me, that this is the true way of administration. Maybe where you work people may think it is too bad that Linux doesn't automount every piece of crap that may either trash the system or give a chance for information leaks. But, on my years of sysadmin I consider that this is one of the best features not only of Linux but of the whole *NIX family.

Re:Is media automount in the kernel yet?

[brunes69]

What are you going off about? How can having a CD automount give way to having "information leaks" (on a read only media?!?) or "trash the system"?? In order to do this you would have to run something on the CD, which is a totally different point of security. You're rambling about nothing.

Re:Is media automount in the kernel yet?

[Junta]

I know mandrake has this feature, at least the one you describe, as 'supermount'. Autofs/Automount is there and has been for a long time, but is less suited for removable devices than it is nfs mounts. Supermount is a far cleaner implementation for local devices. People on kernel-devel would probably argue that the implementation could be done using AutoFS and that would make supermount redundant, but I like supermount.... Between supermount and lpp, you have a decent Desktop OS...

Re:Is media automount in the kernel yet?

[brunes69]

If you think that the rockridge / etc extentions are flakey, then why did you build them into the kernel in the first place? Oh, and if you don't want to use rockridge / etc, then all you have to do is put the override options in your FSTab and the automounter wouldn't use them.

Re:Is media automount in the kernel yet?

[Ektanoor]

Very unfortunate example of yours. The CD is trashed, someone inserts that piece of trash into this super-feature automount. However, this dodo, instead of checking that dumb CD, he goes take some coffee. In seconds, the machine starts to sloooooooooow doooown as the dumbish CD-reader kicks more and more interrupts into the system. True story. I drunk his cup of coffee while he ran away to KICK THAT DAMN CD OUTTA THE SERVER 'CAUSE IT IS STUCK INSIDE!!!

Other example. Mr. A. H. Wannabeahacker has an account in some machine. He inserts the super-pupper CD - CrACkZ, hACKz and SuXXs. Plays a little bit and turns the server into a washing machine. Another true story. Those who work at University computer classes may have seen this a few times...

So people. I know that the autofs features are pretty cool. I do use them. But in user workstations ONLY! The lack of automount in a desktop station is a distro problem not a kernel one. All the basic infrastructure for automount is in the kernel already. However there should be some more tweaking on it, as certain types of ZIPs, CDs and HDDs may seriously influence the performance of the kernel while being mounted. It would be great to see some some kind of double checking of errors so that certain cascades wouldn't happen.

Re:Is media automount in the kernel yet?

[Ektanoor]

A. I am not the only admin in my workplace.
B. The majority of admins comes from old damn good Win9x times.
C. This majority has a tendency to make things MazDie alike.
D. So they put autofs even in some major server.
E. I have can only two hands, two eyes and two ears (well something more but that's superfluous here).
F. I have one head only. And my head is frequently on heatsinker.
G. Certain distros love features. Like giving supermount from start. But if its not there, the guys in B. do everything to put it.

So what do you think? I know... "#brain$ rm -rf nerves ; echo "YOU LAMERZ!!!" ; cat "Why *NIX doesn't autofs from start - 10th edition""

Re:Is media automount in the kernel yet?

[10Ghz]

So why not make it optional?

"Supermount: Enable if used on a desktop-system. Disable if used on a server"

What's the harm there? desktop-users get their ease of use, server-admins get their security and stability.

Reiser4

[KagatoLNX]

I *sooooo* hope the Hans gets off his butt and gets ReiserFS 4 in this one. If you follow the LKML closely (or just read the Kernel Traffic, http://kt.zork.net/kernel-traffic/latest.html) then you may have heard he's sweating a bit on getting it in.

Reiser4 may just revolutionize the way the some people do stuff. I mean, next system I want to be able to do:

# cat /etc/passwd
root:x:0:0:root:/root:/bin/bash
jim: x:100:100:jim:/home/jim:/bin/bash
# cat /etc/passwd/jim/uid
100
# echo /bin/ksh > /etc/passwd/jim/shell
# cat /etc/passwd
root:x:0:0:root:/root:/bin/bash
jim: x:100:100:jim:/home/jim:/bin/ksh

All that *and* have transactional data commits with a very small performance hit!

(ReiserFS Trolls: Go ahead, bring it on!)

Re:Reiser4

[BJH]

He's going to submit it on the 27th. Linus gets back on the 27th, but probably won't actually start looking at pending patches until the 29th or 30th. I'd say it's got a 50-50 chance of getting in to the first 2.6-pre kernel, but an excellent chance of getting in before 2.6 is actually released (don't worry about the Halloween freeze - there hasn't been a major kernel release in living memory that didn't have some enormous chunk of code dropped into it at the last minute).

Re:Reiser4

[KagatoLNX]

Well, perhaps that was a bit mis-stated.

He's supposedly in the first major "cleaning-up" stage with basic stability already reasonably taken care of.

That being said, I think Hans has always been pretty good about limiting what he tries to do to small bite-size chunks, so I don't worry too much.

Re:Reiser4

[IamTheRealMike]

However, I'm not clear what feature of Reiser4 it is that you're demonstrating. Could you provide an explanation to accompany your example?

The feature he's referring to is the ability to treat files also as directories, and it promises to be a majorly cool enhancement to Linux that rewrites one of the most basic assumptions of OS design. The idea is that by improving the power of the filing system layer, and boosting the performance of the FS for small files, the need to have databases layered on top (and even /etc/passwd is a database) is eliminated. This in turn leads to a more powerful OS as that power is made generic, so being available to everything. It's better explained by Reiser himself.

One problem - 2.6/3.0 won't have that ability. What it will have is a special system call reiser4() that Hans can play with. You won't be able to "cat /etc/passwd/mike/group" anytime soon, unfortunately, this kind of major change takes a long time to work its way though the system. The reiser4 call will allow Hans to experiment with the new semantics before we even start to think about merging with the actual kernel.

Why is small file performance so important (this is the area where RFS kicks the ass of, well, pretty much everything else)? Because there are quite a lot of files out there which would actually be better stored as lots of small files. /etc/passwd is one good example, there are others. The reason they aren't currently stored as files is because traditionally filing systems have sucked when you have lots and lots of very small files, and we're talking like perhaps 5 byte files here. Reiser4 has some extremely clever algorithms in it, which mean it's good at small files but also large files too.

Of course, this is just the start of a much bigger picture, that'll see the filing system become something akin to a searchable knowledge store. Unfortunately, it's not going to happen quickly. For starters, if you were to suggest to the maintainer of app foo that they should store their data as lots of small files, they's say "no way, some of my users are on ext3, or xfs, or jfs" etc. Reiser has great vision, but he's not the only player in this field, and I have a nasty suspicion that the goal of exploding out large files into filing system structures could prove to be difficult while other filing systems are prevalent. Let's hope not, eh?

Re:Reiser4

[Effugas]

Of course, this is just the start of a much bigger picture, that'll see the filing system become something akin to a searchable knowledge store. Unfortunately, it's not going to happen quickly. For starters, if you were to suggest to the maintainer of app foo that they should store their data as lots of small files, they's say "no way, some of my users are on ext3, or xfs, or jfs" etc.

ReiserFS in Userspace.

--Dan

Re:Reiser4

[IamTheRealMike]

Read the whitepaper on namesys.com - he'll be introduced set theoretic naming later on, allowing for associative databases (not relational like longhorn).

Re:Reiser4

[peter]

That might work. You wouldn't have one reiserfs volume mounted, like if you were using reiserfs for /home. Each program that wanted to have a config directory of small files could create a config file that was really a database, but was accessed through the reiserfs library like it was a directory. The reiserfs library might include a wrapper for open(2), etc. to do this more transparently. From the kernel's perspective, it would be like a config file with binary data in it. Messing around with the config file with other tools would be a bit of a problem, unless your shell could transparently look inside the reiserfs files and handle redirecting to/from them. You'd probably have to use something other than mv or cp to get files between the "directory" structure of the reiserfs file and the actual FS that the kernel knows about.

Still, it would work if there wasn't much need to deal with the file other than from special applications, which could use a library to see it as a directory. This would let people get started using such structures for config info before reiserfs is widespread.

Re:Reiser4

[peter]

> I think the point is, at least in this example, you want to be able to edit /etc/passwd with vi.
> You wouldn't be able to do that if it were in some form of real database. However, since flat files can represent
> database tables when given a delimiter it shouldn't be difficult or resource intensive to write
> these functions into your filesystem. This is true innovation, I like it. :)

I don't know if I've got this right, but one thing you get from doing this is that you don't need to lock the whole password file to change something in it. You could echo 1001 > /etc/passwd.tmp; mv /etc/passwd.tmp /etc/passwd/peter/uid to change the uid of an account atomically. (Could someone who uses reiserfs's fancy features tell me if that is correct and/or possible?)

Can't help but notice...

[Saint Aardvark]

...that he's left out SearchKing's on-the-fly-Google-re-ranking patch.

Guess we know which kernel guru has started taking $ from Google!

New console layer

[jaymzter]

I for one am totally psyched about re-writing the console sublayer. It's so aesthetically annoying to be running a multi-headed system, yet be reserved to only one head when on a tty. I think this has a high geek factor

this seems a bit premature

[bripeace]

I've been reading the list and well..
This is about 1 of 3 different posts talking about 'what needs to be shown to linus when he gets back'

This is also the very first post of this one thread specific.. theres been about 5 or 6 more major things added to the list that people are hoping to get in ... Linus essentially said he wasn't leaving anyone in charge so they're trying to get one main list to give to linus (with links where possible) so that he can quickly go threw everything when he gets back

Also.. it seems noone on the list is sure whether this will be 3.0 or 2.6 at least noones given any real definate answer as far as I could see..

the lastest version of this list is here.. which compiles all the other threads in one.. is here

Re:Linux 3.0

[Fastball]

You meant to type GNU/Linux, right?

Sincerely,
RMS

Re:Linux 3.0

[cperciva]

Linux ** (just take care of all the letter names at once)

If you only want to take care of two letter tags, shouldn't that be:

Linux ??

Linux version names

[ohboy-sleep]

By the way, Linux 3.0 will officially be known as "Linux III: The Domination" and when they get around to Linux 4.0 it'll be "Linux 4: Citizens on Patrol"

obvious missing patches...

[Lumpy]

Many of you dont like it, but delivering Linux to the masses... the LPP (Linux Progress Patch) is highly important and need to be incorperated into the kernel so that it doesnt become a "left behind" item.

Yes, not seeing all the bootup messages is not highly important... but to a timid user that freaks when the computer beeps it is important. (I agree, people like that need to be kept away from technology... but these people here HAVE to work.)

Linux's acceptance on the desktop needs to have "eye-candy" like this that doesnt lower performance.

Re:Linux 3.0

[Hard_Code]

No, * is for all acronyms. The second star is for fail-over/HA.

agreed

[Ender Ryan]

This one thing is a major hurdle for _some_ new users. My wife in particular gets very annoyed with having to (u)mount disks before using/ejecting them.

Re:agreed

[Lumpy]

ONLY intel users...

anyone that has used a MAC or any other proper system that uses a powered eject or locking media like it is supposed to be so that you dont trash the filesystem on the disc.

Re:agreed

[Iamthefallen]

You mean the very logical method of dragging my floppy to the traschcan? That really doesn't make much sense.

Re:agreed

[red_dragon]

You can just Ctrl-click (or right-click, if using a two-button mouse) on the volume you want to eject, or choose "Eject Disk" from the Special menu (OS <= 9.x). Having said that, though, the first time I used a Mac (Mac Classic running System 6) and wanted to eject the floppy disk, the first thing that came to my mind was to "drag the floppy icon to the trash bin". And it worked.

AOL???

[brunes69]

There are still 7 days till the end of Linus's cruise, but that's not much time to get guinea pigs to publicly pipe up with a hearty "AOL!" of support for your work...

I didn't think a hearty endorsement by AOL would be good news for anything!

Re:AOL???

[bourne]

"AOL!" in this context means "Me Too!" It came about because people associated replies (usenet, web discussion boards) that had "Me Too!" and no useful content with newbies, e.g. AOL users.

Always messes me up. I always think "Army Of Light"...

DOS didn't have automount.

[Inoshiro]

DOS had stateless device access. Until you tried to look at a device, DOS would not touch the device (floppy drive, hd, or CDROM drive). But when you did change to the device, it would try and read in its base directory and bootsector.

Windows emulates its behaviour towards floppy disk drives, as you will find out very painfully if you click on the A: on a computer without a floppy drive (which, for me, is all of them), or without a disk in the drive.

Automount only works on hardware that gives feedback on when media is inserted (such as a CDROM drive). To prevent Badness (TM) in the blocklayer, the automount has traditionally been eschewed in favour of explicit mount. Why? Try removing a CD that's being read from in Win9x, and watch the blue-screen "Please insert CD labelled ..." as the kernel catches a block layer exception. This can be worked around by adding drive locks every time the drive is accessed, but it's generally considered to be a hairy problem best solved by having a smarter user.

Of course, many distributions include the (separate) automount patch anyways, and people who want this behaviour won't be rolling their own kernels any time soon.

Re:DOS didn't have automount.

[brunes69]

Point taken on the blue screen, but how about in Linux when a process running off the CD freezes your console, adn the only way to get around it is to reboot (since you can't just open the CD and have the process die like in windows)? This has happened to me several times in the past. Nont to mention the number of time sI was in windows and wanted to read a serial # off of the cd, so I just open it, blue screen appears, write it down, put it back in, hit enter. No harm done. I construe the opening of cd == death to be a feature, not a detriment.

Re:DOS didn't have automount.

[sg_oneill]

Actually, what's with CD's and intel machines?
It seems the only thing that locks my windows. The whole damn machine just falls asleep while waiting for CD seek. Same with debian. On almost any pc it's the same.
Do OS's traditionally implement these crappily or is it in the implementations in modern ide/mobo/cdrom hardware?

Re:DOS didn't have automount.

[Paul Komarek]

My first action after installing Red Hat, every time:

rpm -e magicdev

-Paul Komarek

Re:DOS didn't have automount.

[FrankNFurter]

You must have a newer XP version than I do. dir a: without a floppy in the drive creates the following message on my system:

Microsoft Windows XP [Version 5.1.2600]

(C) Copyright 1985-2001 Microsoft Corp.

C:\dir a:

The device is not ready.

Re:DOS didn't have automount.

[Paul Komarek]

I've been told it's that the cdrom drivers are typically 16 bit, and that 16 bit processes use the old type of scheduling (non-preemptible).

-Paul Komarek

Re:DOS didn't have automount.

[peter]

> Actually, what's with CD's and intel machines?
It seems the only thing that locks my windows. The whole damn machine just falls asleep while waiting for CD seek. Same with debian. On almost any pc it's the same.

Try running hdparm -u /dev/cdrom. -u tells the kernel IDE driver to unmask interrupts while waiting for things. Presumably, the cdrom driver is waiting for a seek with interrupts disabled. Read hdparm's man page before you use it, and be prepared for a system lockup in case your hardware doesn't like it. (save your files, and preferably shutdown to single user mode.) If things seem ok with concurrent access to the cdrom and the hard drive, run hdparm from your boot scripts. Do test things first, though. Crappy mobo chipsets can cause data corruption when accessing both IDE channels in parallel if you enable DMA on your hard drives. -u is not so risky, wrt. to data corruption. The main risk would be a crash.

However, if your system is going to crash because of IDE driver settings, it won't take all that long to do so. If the system is stable while you dd if=foo of=bar all over the place for a quarter of an hour, you shouldn't have to worry about occasional crashes.

> Do OS's traditionally implement these crappily or is it in the implementations in modern ide/mobo/cdrom hardware?

Linux implements things conservatively because of all the crappy old hardware that might crash with. Someone should make a bootdisk that tests all the things the hardware can do without crashing, and save info on how to enable it normally. Linux 2.4 has an option to enable DMA by default, and I think distros usually enable that these days, since ATA100 is so much better than PIO mode 4!

And don't forget:

[FooBarWidget]

Linux .NET Professional Edition Second Edition Service Pack 3.

LinuxXX

[macdaddy]

Screw LinuxXP. We need to sneak it in the back door of our shops. I think it should be LinuxKY.

All your software are belong to Lucent

[yerricde]

take a look at Plan 9

Lucent's Plan 9 operating system has a bug in its license:

The licenses and rights granted under this Agreement shall terminate automatically if (i) You fail to comply with all of the terms and conditions herein; or (ii) You initiate or participate in any intellectual property action against Original Contributor and/or another Contributor.

This effectively gives the Plan 9 contributors an unlimited non-exclusive license to use any copyright, patent, or trademark belonging to any licensee. For instance, if you install Plan 9, the license gives Lucent the right to violate the GPL on all software that you have developed, to pirate music you've written and recorded, and to pass off its products as yours.

Read more about the bugs in Plan 9

Why aren't Oopses dumped to swap?

[emil]

Dumping the panic to swap seems more sensible than allocating another partition. This is how other OSes do it.

Re:Why aren't Oopses dumped to swap?

[Sheetrock]

I can only guess, but two reasons this might be the case are that the swap doesn't necessarily have to reside in its own partition but can be in a file or simply unused on a system, or (more likely) that this prevents the possibility of obliterating the dump in the process of trying to view it. Also, the software apparently lets you do a 'crash dump' from a live system, which would be inconvenient to have overwrite swap...

This seems like the safest option, because it's isolated from the Linux system at any other point, but it would be nice to get the swap option as well for people who aren't interested in the fancy stuff unless something goes seriously wrong.

What should be embedded into Linux kernel 3.0

[Ektanoor]

Well, this is what some people really wish:

Internet Explorer.
GUI.
The Eternal Flat Desktop for dummies.
Word, Excel, PowerPoint, Solitaire.
Palladin .NET
WYSINAWYG
WYGINAWYW
Easter Eggs
Make desktop "user ready". Forget the flame.
Forget the bugs, claim the features.
Add 100Kb EULA into the kernel itself.
Sell it and yourself to Bill Gates.
Rename it to Windows.
Sell it for $400 and threaten everyone who will not follow you.
Write a small text, anonymously authored - "Why I switched from Linux to Windows" and claim how your customers are deeply satisfied.

More complete list

[awptic]

This page contains a complete list of every new feature that has gone into 2.5, and other features waiting to be integrated and their status:

http://kernelnewbies.org/status/latest.html

Re:Let me be the first to come out with the bad jo

[GreyWolf3000]

Here is the stupid thread you're looking for. Good day, sir.

Re:Linux 3.0

[den_erpel]

No, this time Linux is correct since the topic is about the kernel.

If you are talking about the Operating System, you should address it as GNU/Linux (same as you have GNU/Mach).

e.g. Debian, SuSE, Redhat, ... distribute GNU/Linux (you would not do much with a Linux distribution).

You must have amnesia RMS, since you learnt us to cite:
GNU is the operating system and Linux is one of its kernels

Re:OK, that's it, I'm 'switching'

[rastachops]

Id say try out the Mandrake distribution (Linux-Mandrake.com) as its very good for making the hop from Windows to GNU/Linux.

Ive tried Red Hat but personally prefer Mandrake 9.0. Checkout Linux.com & for any hints tips / tutorials Google is always the best bet.

Re:Already got a beta version....

[CvD]

Snipped from Linux Kernel 2.5 Status:

o in 2.5.31 Support insane number of processes (Linus Torvalds)

Just how insane would this be? Anyone know?

Re:Let me be the first to come out with the bad jo

[stephend]

Windows only started taking off when it got to version 3.1. Maybe the same is true of Linux?

Re:OK, that's it, I'm 'switching'

[IamTheRealMike]

I'm going to buy a new system in a month or so, and a stable linux 3.0 (I've heard the odd/even thing...3.0 is divisible to 1.5...right?) seems like a good excuse to finally install linux.

Welllll, if it's 3.0 that you want to try, I'd wait a few months (at least). For starters, most of the major commercial distros (and for a beginner, you want a commercial distro) have just had major revs and so won't be upgrading their kernels for a while. Linux 3 will a lot of testing before all the wrinkles are ironed out, even after release, so when you start seeing companies like RedHat, Mandrake and SuSE (the one I use) shipping with a 2.6/3.0 based kernel, then you know it'll be ready.

On the other hand, if this story just piqued your interest, and you want to try anyway (2.4.18 is pretty good) then head over to linux.com where they have a good newbie article and a ton of links to help you out. A good boxed commercial distro will come with printed manuals to get you started, and if you get stuck, want to know a command etc (and you will) come say hi to us all on irc://irc.freenode.net/#linuxhelp

Linux is what you want it to be

[siskbc]

when people start complaining about how hard it is to install and configure. Because in the end it seems these people don't realize anymore that they are complaining about Linux itself.

What a bunch of garbage. Who the hell said that Linux is supposed to be hard? Why? So you can be an 3!337 hacker when you get it installed? That attitude is why linux doesn't take off on the desktop - because when newbies interested in it look for help, all they hear is "RTFM." Yeah, those manuals are real easy to read.

You seem to be under the mistaken impression that linux is a religion, and that we should all be bowing down to the Great God Penguin. Crap on that. Me personally, I'm not looking for an OS to become my hobby. I'm not looking to spend hours installing a damned window manager.

All I'm looking for is an OS that works, that I can customize aspects of if I like, that comes with a good compiler, that doesn't crash, and that isn't too bloated. Linux fits all of these requirements. I personally use Slackware, which was because I wanted to learn some about the guts of Linux, but I also wanted intelligent default configurations.

However, none of this gives me, you, or anyone else the right to insult someone because they don't want to spend hours, or days, on an install. Some people have jobs, and social lives, that together preclude spending such time on an OS. For people who want an OS that just WORKS and gives them the flexibility to do what they nead to do, a distro like Mandrake may make sense. This doesn't mean that they're less "linux" than you.

I think some people need to re-evaluate why they use linux in the first place. Is it because they are using it as a replacement for a social life? A replacement for religion? An outlet for M$ hatred? Because they like hacking open OS's? Or because it just does what they want it to? Personally, I'm of the opinion that the first three groups can just piss off.

Prepare to reboot your box next year...

[DocSnyder]

How could my server ever reach 1000 days of uptime with Linus throwing out new major kernel releases every two years? ;-)

What about ACPI support?

[MtViewGuy]

I've heard one of the goals with the next major Linux kernel is Advanced Configuration and Power Interface (ACPI) support for hot-docking of external devices and hot-swapping of devices on certain types of PCI slots.

With automatic hardware detection and configuration of peripherals in a standardized manner, maybe it might convince peripheral manufacturers to write Linux drivers en masse.

Already has this feature, has had it for years.

[Inoshiro]

Go look at http://members.aol.com/axcel216/lastweek.htm:

"MessageBackColor=8 To specify the BSOD (Blue Screen Of Death) background (screen) color. Default is blue (1). See "BLUE (OR ANY OTHER COLOR) SCREEN OF DEATH", also in TIPS95.TXT [part of W95-11D.ZIP], or in MYTIPS31.TXT [part of W31-11D.ZIP], for complete details.

MessageTextColor=C To specify the BSOD (Blue Screen Of Death) foreground (text) color. Default is bright white (F). See "BLUE (OR ANY OTHER COLOR) SCREEN OF DEATH", also in TIPS95.TXT [part of W95-11D.ZIP], or in MYTIPS31.TXT [part of W31-11D.ZIP], for complete details."

Oh well, nice attempt ;-)

Re:Already has this feature, has had it for years.

[grammar fascist]

So...what do I set MessageBackColor to if I want the Paisley screen of death? Or streaming video?

Oh well, nice attempt. ;)

They were just making a point

[Zemran]

... :)

Your system must have something mine doesn't.

[Inoshiro]

Wintendo has no floppy device, and I have disabled the floppy controller in the BIOS. Accidently clicking it in explorer makes the A750 into a paperweight for a few minutes, as I have to deal with Windows trying to work with what isn't there. It's probably partially the BIOS's fault, but the entire thing is a cluster fuck that should've been fixed back in the days of Win95.

Re:Your system must have something mine doesn't.

[Inoshiro]

Windows "finds" it again and again.

Trust me. I've looked. There are a lot of BIOSes which "show" a floppy drive + controller to the OS, despite one not being there, and other settings being set as such. Because Win9x would crash hard if there was no floppy controller + drive in many common configurations.

A proper RegEx

[Plutor]

'Linux [A-Z]{2}'
or
'Linux [[:alpha:]]{2}'
or even more flexibly
'Linux ..'

You should be able to starting with 3.0 (2.6)

[alder]

The kexec patch should do the main part of the trick. And its status is Ready.

Re:Not quite

[ShavenYak]

Anyway, he actually went back to the mainland in 1975 and lived for 11 more years after that, during which time several of his books (including "battlelield earth") were written, so i don't think there's much question whether he was alive during the time on the boat.

If Battlefield Earth were attributed to me, I'd want people to think I had actually been dead before it was written!

Re:We Want Linux 4.0

[WebMasterJoe]

When engineering decisions are made for marketing reasons, Bad Things happen.

When any decisions are made for marketing reasons, Bad Things happen!

Wow!

[theLOUDroom]

The kexec patch [xmission.com] should do the main part of the trick. And its status is Ready [kernelnewbies.org].

Holy crap that's sweet!
I can upgrade my kernel without rebooting!
I had no idea you could do this. Yet another reason linux rocks.

Mandrake

[Dave_bsr]

Mandrake runs fine on my pentium 166. the graphical installer is slow...and the text installer is just fine. Just install the minimum packages - and you have a nice, sweet system with no X for 100MB. Yes, that's mandrake. The "big, bloated, ugly-but-easy-to-install" mandrake. *shrug*.

Ever heard of /dev and suid?

[KPU]

Consider a CD with Rock Ridge that has device files (/dev/hda comes to mind) with ownership to a non-root user. A user can get around /dev/ permissions by doing this if it's mounted without nodev. Also consider a suid executable owned by root. A user can run it and get root privs unless the nosuid option is set. When one user can mount as another user (as supermount would do) or with the wrong options, that's a security hole. Secondly, suppose root is copying stuff off a CD or wants to access it remotely. Do we let joe user umount it by pressing eject?

Re:Linux 3.0

[Hitokage_Nishino]

No, only Debian distributes GNU/Linux. SuSE distributes SuSE Linux, and Red Hat distributes Red Hat Linux.

I'd remark on how the kernel increasing influence on userland makes the notion of "GNU/Linux" obsolete... but I'll leave it at that to minimize flames. ;)

Re:Interesting cruise by the way

[tpv]

Actually Linus is listed as a speaker on the front page, but his bio isn't on the speakers page.

Perhaps they thought he need no introduction.
Or maybe he just signed up as an attendee and they roped him into speaking.

Re:Linux 3.0

[civilizedINTENSITY]

Yeah right. So rm everything but your kernel and init scripts and see what you can do with your box. Not much.

Re:Linux 3.0

[delta407]

No, if you're using regular expressions (you're on *nix, right?), that'd be:

Linux ..

Doesn't quite have the same ring, though.

UML?

[autopr0n]

What exactly do you mean by that? Surely not Uniform Modling Language. :P

Uh...

[autopr0n]

We are talking about automount, not autoplay...

LinuxYQ?

[achurch]

• Linux, You Queer!
• Linux Yields Quietly
• Linux, Yummy & Quick
• ...

Linux 3

[hayriye]

Ok, take Sun's route for Java, name it as Linux 3 (or III) and continue to maintain it as 2.6 vice versa...

Re:2.4.x still compiles on mine too, (in 3 hours)

[Shanep]

I wrote a script to log the start & end times and let it run.

Why bother? Just use time. You won't have to do any math either!

UDI

[psamuels]

As a device driver developer, do you think Linux's interal APIs change too frequently compared to other Unix operating systems? Would there be much value in creating a more stable, possibly cross-platform driver API for Linux? How much cross-platform driver code can be shared across Solaris, HP-UX, and Linux?

I'm not a driver developer, so I see the question wasn't aimed at me, but this has been done already - the infamous UDI (uniform driver interface) project, brainchild of Intel three or four years ago if I remember correctly. It was actually a uniform ABI, not just API. And it was endorsed by most of the commercial x86 Unix vendors. I guess they were hoping they could leverage Linux's comparatively broad driver support. (:

A few in the Linux kernel community, such as david parsons, supported UDI ("let's start harvesting Unixware drivers", I believe he said), but most of the others, including Linus, thought it was an idiotic waste of their time to support a third-party stable API (not to mention ABI) that would benefit all the other Unix vendors a great deal more than it would benefit Linux. Linus really doesn't like having his hands tied by in-kernel API compatibility. And, unlike most other OSes, Linux doesn't have much of a community of third-party driver developers who don't contribute directly to the master source, so Linus has the rare luxury of being able to change the API whenever a technical reason for doing so comes up. Pity the poor developers of AIX, Solaris or Windows, who have to hold on to any badly-designed APIs until the end of time so third-party video drivers (Windows) or those Oracle-specific kernel features (Unix) don't suddenly break.

(Note that the userspace API is different and is held near-sacred - nobody wants to break applications for no reason. Certain specialised apps like "ps" and "lsmod" can expect a rough ride, API-wise, but for the most part, you can still run 10-year-old Linux code.)

I think UDI eventually died the slow death of apathy - apparently without all those Linux drivers to leverage, there wasn't much point. But I could be wrong; perhaps Unixware and Solaris/x86 still support it.

Packet-writing?

[10Ghz]

According to kernelnewbies.org, CD-R/RW packet-writing is in beta. But it has been for a while now. Any news on that front? Does it work?

OT: multi-{headed,keyb} sys (Re:New console layer)

[alfaiomega]

It's so aesthetically annoying to be running a multi-headed system, yet be reserved to only one head when on a tty.

Will it be easily possible (or maybe it is now, then excuse my ignorance) to have e.g. a system with 4 monitors, 4 keyboards and 4 mice to act like a server and 4 diskless terminals, only cheaper? I suppose it'd be easier to have few text consoles then few X servers, but I have really no idea. It would be cool to have graphic cards with mouse and keyboard sockets in them, but I don't think there's such a thing, at least I've never found anything like that.

How about RAID-0 that gives a performance gain?

[Shanep]

Back when I was using 2.2 kernels, Linux software RAID-0 was giving me 30% or more transfer rate increases. I was kinda hoping for something in the 80%+ area but anyway.

Linux 2.4 has been giving me transfer rates that are slower than a single drive! Since I've pretty much moved completely over to the BSD's for many months now, can anyone tell me if software RAID-0 performance under Linux after 2.2 is now decent?

Re:Self Righteous morons

[nurb432]

BSD