Linux 3.0 Will Be Faster Than 2.6.39

sfcrazy writes "While we were thinking that the announcement of 3.x branch was nothing more than Linus' mood swing, it seems there is more to it. Linus wrote on the Linux Kernel Mailing List, '3.0 will still be noticeably faster than 2.6.39 due to the other changes made (ie the read-ahead), so yes, the regression itself is fixed.'"

Comments

Faster?

[ZombieBraintrust]

What does faster mean? What will be faster? Are they talking huge Linux servers or Linux Desktops? Latency? User Interface?

Re:Faster?

[ZombieBraintrust]

Kernel optimizations could result in User Interface improvements. Or it could help batch jobs at the expense of the User Interface.

Re:Faster?

[TarMil]

Well, the subject of the news is the kernel, not the rest of the OS.

Re:Faster?

[Anonymous Coward]

Pay attention to context. This article is about the kernel. You made a mistake, and it doesn't make them a pedantic troll to point out that you got the context wrong.

Re:Faster?

[DirePickle]

Obviously it means that Javascript will execute faster! Javascript is the alpha and the omega!

Re:Faster?

[shaitand]

Actually the entire OS is the kernel. It's windows and macites who push to redefine the OS as being the entire distribution.

Re:Faster?

[garyebickford]

Whoof, we're going to get into the semantics of what an OS is here! My reference to RMS was not so much about discretion, but by his reputation for enthusiasm for discussing the topic. :)

However I will refer to this quote, from Wikipedia to justify my stance that the kernel is not the OS, but a part of the OS. I will add that I've used many operating systems over the years, and the 'OS' has always referred to the complete package - kernel, core libraries, userland applications, IO and other hardware drivers, etc.

GNU/Linux is a term promoted by the Free Software Foundation (FSF), its founder Richard Stallman, and its supporters, for operating systems that include GNU software and the Linux kernel.[1] The FSF argues for the term GNU/Linux because GNU was a longstanding project to develop a free operating system, of which they say the kernel was the last missing piece.[1] ...
Torvalds wrote, "Sadly, a kernel by itself gets you nowhere [...] Most of the tools used with linux are GNU software."[20] Torvalds also wrote during the 1992 Tanenbaum-Torvalds debate that, "As has been noted (not only by me), the linux kernel is a miniscule part of a complete system".[21]

Re:Faster?

[shaitand]

"I will add that I've used many operating systems over the years, and the 'OS' has always referred to the complete package"

I wouldn't dare to suggest that computer science and not user perception is the best place to define an operating system.

Feel free to peruse this gem http://www.amazon.com/Operating-Systems-Design-Implementation-Second/dp/0136386776 . It is computer science coursework that explains how to write an operating system... err kernel as you say.

As it happens, the MINIX operating system... err kernel, was used as the basis for the book. Once upon a time, Linus Torvalds set out to write an operating system and used this material as reference.

The same author wrote another book, "A History of Operating Systems". You need the history to understand. In the early days all computer operators were programmers and the operating system provided an abstraction between manually controlling the individual hardware components with directly input binary. The drivers, boot system, and memory management were clearly part of that abstraction but since programs weren't even stored but were input one off there was no such thing as a userland program being part of the operating system!

Later storage became more common and some operating systems included some additional helpful applications to make writing your programs easier and later yet pre-written programs began to spread. In some cases like the later dos and apple systems the operating system wasn't even available separately so it became common to refer to the entire operating system distribution as an operating system for short. Many less informed (which ultimately was most) users didn't even realize it was a shortened term.

Re:Faster?

[garyebickford]

Yep, I was there (almost) - some of my first programs were written in IBM 1130 Assembler. :) The machine had 16Kx16bit core, a 1 MB single-platter disk with a one second mean access time. I managed to thrash the poor beast once by nesting too many macros. The key fact is that the control program (pretty close to what we now call a kernel) was on the outside of the disk, the macro-assembler was in the middle, and the user programs were on the inside (or vice versa - I don't recall now). With only 16K of memory, everything the machine did had to be overlaid - except for about 200 bytes or so (I don't recall the number) of code that stayed resident during a job, that basically just knew how to get the next piece off the disk. That could now be called a very primitive kernel.

I suppose this could be considered equivalent in some ways to a bootstrap loader except it continued bootstrapping the various pieces in throughout the process of running a job. Every piece of code had to be loaded over the previous piece in order to run, and each time what we would now call the machine state had to be written to disk. So for each macro call, the machine had to swap bits of kernel, assembler and user code in and out, moving from the inside, to the outside, to the middle, to the outside, to the inside, etc., rinse & repeat. With a one-second access time the 15 minute maximum run time was exceeded before the assembler even finished assembling my 10 or 15 punched cards into machine code. It was a very compact program, but I never did get to run it in its full macro-bedecked glory. I had to turn the program into 100 or so cards of non-macrofied assembler.

I also (much later) had the fun of entering entire programs into an early microcomputer by flipping front panel switches, pushing the 'step' button, flipping panel switches, etc. - make one mistake, push 'reset' and start over. Seymour Cray, when he was still at Control Data Corporation (CDC) was famous for being able to enter the entire 6000 word control program into the early CDC machines from memory using the front panel switches.

So I would say that until we started getting into time-sharing and such complexities, the idea of a kernel wasn't really relevant - there was little or nothing resident in the computer's memory. I think I could safely say that is primarily what a kernel does in a modern multitasking system - provides the environment by which tasks can move safely and efficiently through the system. And the operating system includes all those non-kernel tasks, such as accounting, access control, logging, the many utilities required to provide everything from I/O to temperature control.

Just to put a stamp on this, Wikipedia on Kernels:

In computing, the kernel is the central component of most computer operating systems; it is a bridge between applications and the actual data processing done at the hardware level. The kernel's responsibilities include managing the system's resources (the communication between hardware and software components).[1] Usually as a basic component of an operating system, a kernel can provide the lowest-level abstraction layer for the resources (especially processors and I/O devices) that application software must control to perform its function. It typically makes these facilities available to application processes through inter-process communication mechanisms and system calls.

Operating system tasks are done differently by different kernels, depending on their design and implementation. While monolithic kernels execute all the operating system code in the same address space to increase the performance of the system, microkernels run most of the operating system services in user space as servers, aiming to improve maintainability and modularity of the operating system.[2] A range of possibilities exists between these two extremes.

Re:Faster?

[toadlife]

Good luck operating your system with just the kernel.

Re:Faster?

[MichaelSmith]

You don't browse the web with sysctl?

Re:Well of course! Just like Firefox

[FooBarWidget]

Your cynical undertone that software only get slower and slower is not true. Firefox 4 *is* faster than Firefox 3. Haven't used Firefox 5 yet. Phusion Passenger 3 is 50% faster than Phusion Passenger 2.

Re:good riddance regression

[Anonymous Coward]

good thing the regression is sorted

Even the regressions are much faster in 3.0!

Re:Well of course! Just like Firefox

[billcopc]

Well, Windows 7 *is* much faster than Vista.

Sarcasm fail.

Re:linux 3.0

[Kufat]

They're currently on 3.0 RC4. So I imagine that what will and won't be in the release has pretty much solidified by this point.

prefetch()

[Anonymous Coward]

According to LWN article about removing prefetch, the linux kernel 3.0.0 will have a bunch of prefetch() calls removed from the kernel.

Apparently they were supposed to provide hints to the CPU to prefetch the next item in linked lists, but the hardware does a superior job of it without the hints. Especially in the case of the next item being NULL, which was the majority of the cases.

A very small speedup to be sure, but it's not like there are many low hanging huge wins left.

Re:prefetch()

[ZombieBraintrust]

So it should be slightly faster on new chips. Slightly slower on outdated or specialized embedded chips. (No big deal they can use an version or a custom kernel)

Re:prefetch()

[Annirak]

If this is the case, wouldn't CONFIG_USE_PREFETCH be a better solution?

Re:prefetch()

[blair1q]

It ought to be controlled by a CPU #define, if it makes that much difference.

Re:It should have compelling features

[aardvarkjoe]

A major version should always have compelling features or a shift in model such as drivers.

The Linux development model no longer makes that a useful way to designate version numbers. Why should we be so dead-set on the tradition of version numbers that we can't even break out of that mold when it's useful to do so?

Re:Well of course! Just like Firefox

[Lennie]

Getting past Vista isn't a big achievement though.

Re:It should have compelling features

[badboy_tw2002]

Because some people on the autistic spectrum react badly to changes in established patterns. Remember when Rainman didn't get to watch Whopner? Same thing here. It doesn't make a lot of sense to most folks, but it helps to see it from their viewpoint.

Re:Windows 8

[Annirak]

Windows Vista was slower than Windows XP.

You win some, you lose some.

Re:Well of course! Just like Firefox

[blair1q]

The second cake is even more delicious than the first.

Obviously

[vga_init]

What with the shorter version number, the kernel should now load faster, use less memory, and execute more quickly.

Re:Well of course! Just like Firefox

[sortius_nod]

True, but moving us die hards past XP was a big challenge. I use 7 now on my win box (for gaming).

Re:Marketing

[3vi1]

Not marketing. They very well could have been saying "2.6.40" will run faster.

I love Linux, but the post 2.6.38 kernels have developed in a way as to be completely random as to whether or not they will successfully boot on my x58/i980x motherboard/CPU. All kinds of breakage and improvements hit in the 2.6.39-40 cycle that are going to take a while to even out. So, don't expect the exact same experience as with the 2.6.38 kernel.

Re:Windows 8

[westlake]

Windows Vista was slower than Windows XP.

When, where, why, and by how much?

Re:Windows 8

[aeoo]

On shitty hardware, yes.

When you compare two versions of the same operating system to determine which version is faster, you always use the exact same hardware configuration for both.

Re:Windows 8

[Osgeld]

problem was good hardware for vista came out 3 years too late

Re:Marketing

[DrXym]

Most of the problems Flash has had with non Windows platforms are inherent to the platform. e.g. Flash wants to decode video in hardware, convert to RGB, compose with other RGB elements, and present this preferably in a widget / window running inside a 3rd party app. If the OS / browser impedes any of this Flash will fallback on slower methods and performance takes a dump.

LFW

[Anonymous Coward]

What about LINUX 3.11 for Workgroups ?

Re:Marketing

[Kjella]

Want proof? proof that nobody here can deny? here you go...why does Dell, one of the largest OEMs on the planet, have to disable the repos on every. single. Ubuntu machine they sell and deal with the hassle and expense of running their own repo, even for a small subset of the hardware they sell? Why because if they don't Linux breaks drivers oh fun oh joy!

First of all, if you go to any OEMs website you will see that they usually offer plenty driver downloads, they're not leaving it all to Microsoft. So what you're complaining about is exactly the same practice they have on Windows. Secondly, Microsoft offers their prerelease patches and service packs to OEMs for compatibility testing before they arrive on Windows update, which Linux doesn't. So if Dell was to provide the same level of compatibility testing as they do on Windows, they have to redirect it to their own repos.

Sadly most the stuff that breaks is not in kernel space - Linus runs a pretty tight ship with very low tolerance for regressions. To take the whole PulseAudio mess for example, where is it? Userspace. If you have a problem with any USB-connected device, it's userspace as all basic USB I/O works fine - the rest is in userspace. The bluetooth daemon is mostly userspace. Wireless may be kernel space, but just as often the userspace init scripts and tools are fucked. Not saying the problem isn't real, but that it is mostly a problem with the distribution. That also means the choice of distribution will greatly influence your experience.

Oh and PLEASE don't say LTS, as long as Linux software is tied to the kernel you might as well call that an out of date unpatched OS, because that is what it is.

No, it's an out of date patched OS, so your customers don't get rooted. Enable the backports repo and a lot of the core software will get updated too, rather conservatively of course. Tell me, does the Windows PCs you sell update themselves from Vista to Win7 or Office 2007 to Office 2010 when it's out? Or does it for the most part just provide security patches, like an LTS release? Speaking of which, Office comes with a new version every 2-3 years. Does it then kill you to ship a 1-2 year old version of OpenOffice? Because even an LTS release upgrades every 2 years.

I can think of pretty many other arguments for why Linux isn't taking off, but I don't think yours were all that compelling. Like you say, Dell has a pretty good system for this and if that's all it would take the YotLD would have come and gone long ago. It doesn't have the entertainment apps (games, bluray playback, many streaming solution depends on Windows & DRM) for the home user nor Outlook/Office for the business user. And every app is different from what you know.