Linux 3.0 Will Be Faster Than 2.6.39 179
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.'"
Faster? (Score:3, Insightful)
Re: (Score:3)
it will be faster on Linus' laptop of course (Score:2)
Adoption (Score:2)
Faster Linux adption, of course! :)
Re: (Score:2)
Re: (Score:2)
Sure they do. See /usr/man/man2 ... or wherever your distro has moved the freaking man pages to, because this RH5 I'm on hid the fucking things in /usr/share/man/en...
Re: (Score:3)
You don't browse the web with sysctl?
Re: (Score:3, Insightful)
Re: (Score:2, Informative)
Re: (Score:2, Flamebait)
Strictly speaking, Linux is the kernel - the 'entire OS' is properly GNU/Linux. Most of the core applications and libraries are from GNU. Ask Richard Stallman, he'll provide a couple of hours of instruction this topic, I'm sure.
Re: (Score:2)
Actually in the case of a Macro kernel, the entire OS is the kernel and it is properly referred to as Linux. GNU makes some libs and userspace apps that you can run on the Linux OS though and many distributions package them together... those are properly referred to by the name of the distribution, which may or may not include the word Linux and may or may not include GNU. None of which is at the discretion of Stallman.
Re: (Score:3)
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 [wikipedia.org] 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 drive
Re:Faster? (Score:5, Interesting)
"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 [amazon.com] . 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? (Score:5, Informative)
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 [wikipedia.org]:
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: (Score:2)
Clearly we are going to have to agree to disagree on this point. But really, you need to stop pointing at Wikipedia like its a reference.
A Wikipedia quote doesn't put a stamp on anything, it isn't any better than a random Slashdot post. In fact, I could edit a Wikipedia article and post it right now as a counter-point.
Re: (Score:2)
He's right, a kernel is not an operating system. For instance, consider the Unix OS. There is no "Unix" kernel, there are only compatible implementations of the Unix operating system interface, which is defined entirely at user-level. Some of the Unix interface may be provided in the kernel, like Linux, Solaris, AIX, and some implementations are purely in userland, like Unix binary compatibility layers for microkernel operating systems like KeyKOS, L4, and Minix.
The kernel merely forms the core part of an o
Re: (Score:2)
Lets give it a true stamp:
http://en.wikipedia.org/wiki/History_of_operating_systems [wikipedia.org]
"Operating systems (OSes) provide a set of functions needed and used by most application programs on a computer, and the linkages needed to control and synchronize computer hardware. On the first computers, with no operating system, every program needed the full hardware specification to run correctly and perform standard tasks, and its own drivers for peripheral devices like printers and punched paper card readers. The growi
Re: (Score:2)
Wouldn't you draw a distinction between self-hosted and externally-hosted operating systems? To me the whole distribution is the OS, or at least the parts that you need to rebuild the distribution, specifically because it is possible of doing this.
Re: (Score:2)
"or at least the parts that you need to rebuild the distribution"
The "system" would be the hardware platform. The kernel operates the hardware platform and therefore the system. Everything else operates via the kernel, even the low level libraries if they operate the hardware it is only as the kernel has exposed said hardware.
"Wouldn't you draw a distinction between self-hosted and externally-hosted operating systems?"
What difference does it make if the OS is loaded off a disk via a short path of copper or
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Pretty sure that's what I said?
Re: (Score:2)
Actually in the case of a Macro kernel, the entire OS is the kernel and it is properly referred to as Linux.
I'm pretty sure it isn't.
Re: (Score:2)
This is not correct. The Linux kernel defines hardware abstractions, and you have to invoke its services using system calls by placing certain values in hardware registers, and certain values on the stack, and executing an assembly instruction to transfer control to the kernel. But apps never actually do this, they instead call the kernel services via a C library which does this for them. The Linux op
Re: (Score:3)
Actually the entire OS is the kernel. It's windows and macites who push to redefine the OS as being the entire distribution.
Re: (Score:2)
My point is that it doesn't really matter in the long run.
People will call it what they will. Yelling at them that they are calling it the wrong thing will only make you look like an anti-social geek. Its like the people who refer to their tower as the cpu, or the ones who refer to the hard drive as memory. You can correct them all day long, it doesnt matter though. If there are enough people then eventually the term gets redefined. Language evolves, not much a few people can do about it.
Re: (Score:2)
From now on I will use (Hacker, cracker, script-kiddie), (virus, worm, trojan), (kernel, OS, windows) all interchangeably. I even think I am going to believe in a year zero too.
No really, I give up.
Re: (Score:2)
You don't spit into the wind. Why not its pointless.
I get your sarcasm, all I'm saying is that people can get mad all they want, its just not going to make much of a difference. Feel free to continue to correct people if you wish. Good luck.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
In this thread Linux means the Kernel, period. (Score:2)
ROTLMAO. The title of this article is "Linux 3.0 Will Be Faster Than 2.6.39. This means that in the context of this topic Linux refers to the kernel and only the kernel. Actually understanding the context of a conversation and remaining consistent with it is not pedantic trolling.
linux 3.0 (Score:2)
please tell me he isn't thinking about adopting firefox and chrome's release model...
in all seriousness, it still looks like this is more of a rumor than anything that is going to be done for a while.
Re:linux 3.0 (Score:4, Informative)
They're currently on 3.0 RC4. [kernel.org] So I imagine that what will and won't be in the release has pretty much solidified by this point.
Re: (Score:2)
To be fair, they've been introducing some pretty big changes in 2.6.38 and 39, enough that I am reluctant to upgrade from 2.6.36 on any servers because it's just been changing TOO quickly and I'm afraid of new bugs.
Re: (Score:2)
It marks the 20th anniversary of Linux. It really isn't that hard to understand wanting a major release to mark it.
Re: (Score:2)
The kernel has long been on a quarterly release schedule. One month merge window, two months of weekly RCs. If Linus is feeling very generous maybe you can sneak into rc1 or rc2 but don't count on it. So every feature is known roughly two months before release, unless it's backed out because it causes too many regressions. So it's actually Firefox that is changing to releases every 3 months, even though they call them 4, 5, 6, 7 while the kernel will be 3.0, 3.1, 3.2, 3.3... just like they've been 2.6.35, 3
Re: (Score:2)
Hell, he coulda gone to 2.7 if he was tired of feeling old.
prefetch() (Score:5, Informative)
According to LWN article about removing prefetch [lwn.net], 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: (Score:3)
Re:prefetch() (Score:4, Insightful)
If this is the case, wouldn't CONFIG_USE_PREFETCH be a better solution?
Re: (Score:2)
If this is the case, wouldn't CONFIG_USE_PREFETCH be a better solution?
It seems likely that any CPU which has the prefetch instruction also does hardware pre-fetching.
Lack of hard evidence (Score:2)
I'm going to parrot what ds2horner said on LWN - there needs to be evidence that the prefetching is still a benefit to somebody [lwn.net]. If not enough people come forward with this then it's a maintenance burden for an uncertain benefit (for what it's worth it is apparently a win to keep PREFETCH on K7s).
Re: (Score:3)
It ought to be controlled by a CPU #define, if it makes that much difference.
Obviously (Score:4, Funny)
What with the shorter version number, the kernel should now load faster, use less memory, and execute more quickly.
Re: (Score:2)
Totally, every single time a system call that returns the version string is run it will load faster, use less memory, and execute more quickly!
LFW (Score:3, Funny)
What about LINUX 3.11 for Workgroups ?
Re: (Score:2)
Re: (Score:3)
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 [modrails.com] is 50% faster [phusion.nl] than Phusion Passenger 2.
Re: (Score:3)
The second cake is even more delicious than the first.
Re: (Score:2)
Firefox 3 is faster than Firefox 2. Firefox 4 is faster than Firefox 3. You do the math. If you don't believe, download Firefox 2 (don't forget to clear Firefox's caches and history) and compare it for yourself.
Re: (Score:2)
Where are these "most benchmarks"?
Re: (Score:3)
Well, Windows 7 *is* much faster than Vista.
Sarcasm fail.
Re: (Score:3)
Getting past Vista isn't a big achievement though.
Re: (Score:3)
True, but moving us die hards past XP was a big challenge. I use 7 now on my win box (for gaming).
Re: (Score:2)
Especially since Windows 7 is just a vista service pack...
Windows 7 is a Vista service pack in the same way that XP was a service pack for 2000 - there were significant visual and under-the-hood changes. The visual changes were more noticible in XP since they introduced themes, but if you turned them off, XP looked about as similar to 2000 as 7 does to Vista.
In other words, they're more than just a service pack. Granted, not much more, but your statement is an oversimplification. In the case of Windows 7, it made an unusable OS something that didn't make you wa
Re: (Score:2)
there still people who care about speed ? the same ones that care about dick size ?
Mine are both are "good enough, now go do something interesting with it"
Something interesting... (Score:2)
Playing videos on my laptop without killing battery life and burning my lap would be interesting. Granted, that's Flash's fault, not the kernel's fault, but still.
But some examples:
I just played Angry Birds in Chrome -- I don't have a smartphone. Before JavaScript got fast (thank you, v8!), the choices would've been Flash (which is even slower on Linux than elsewhere, and 32-bit), Silverlight (really only works well on Windows and Mac, other OSes get screwed), or a downloadable binary (more work to port, so
Re: (Score:2)
Youtube isn't "the same set of features" as a DVD. Granted, there's no reason it should require as much more than a DVD as it does, but it certainly should require more.
Re: (Score:2)
How much of this has anything to do with YouTube? I think YT uses the DRM on actual movies, but on individual videos, I can trivially capture the stream -- and often, I can use HTML5 instead.
Also, how much of this has anything to do with politics? It's my understanding that it has much more to do with the demands of content owners, rather than politicians.
Re: (Score:2)
That works, but the simplest mechanism I've found is just to watch the HTTP traffic. Chrome makes this easy -- open dev tools, refresh the page, watch Flash try to stream that video, copy the URL, and often wget will work. (At which point, I close the Flash player to save bandwidth.
Re: (Score:2)
The problem with ARM, is that many people are locked into closed source applications which have only ever been compiled for x86... ARM is great when you have sourcecode and can compile your applications for it.
Re: (Score:2)
Re: (Score:2)
Perhaps you should learn to read. They has some regressions in performance from a specific kernel version forward. They have a fix for the regression; well, mostly so. Thusly, when the fix is released in the 3.x, Linux version 3.x will be faster than the window of Linux kernel versions in which the regression appeared. Saying Linux 3.x will be faster than 2.x is completely false and even acknowledged in the original Linus comment.
Re: (Score:2, Funny)
good thing the regression is sorted
Even the regressions are much faster in 3.0!
Re: (Score:3)
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: (Score:2)
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: (Score:2)
Wapner. Definitely, definitely Wapner.
Re: (Score:2)
No, it should mean whatever the Linus decides it means. It is his to do with as he likes. If you want you can fork Linux into Bluenix and use your version numbers.
Re: (Score:2)
Re: (Score:2)
A major version should show up every six months with whatever we feel like dropping into it at that point.
Signed,
Mark Shuttleworth
Re:Windows 8 (Score:5, Insightful)
Windows Vista was slower than Windows XP.
You win some, you lose some.
Re: (Score:3)
Windows Vista was slower than Windows XP.
When, where, why, and by how much?
Re:Windows 8 (Score:4, Insightful)
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: (Score:2)
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.
Yes. With modern hardware, Vista is (usually) faster than XP on the same hardware. Windows 7, more so. On old (or very cheap) hardware, XP is (usually) faster than Vista. Result vary between components and benchmarks though.
Re: (Score:2)
Since the vista service pack, windows 7, runs nearly as fast as XP when given enough hardware I guess some of us would have expected at least that level of performance at initial release.
Re: (Score:3)
problem was good hardware for vista came out 3 years too late
Re: (Score:2)
You know its bloated when even
ls -la /usr/bin/vim /usr/bin/vim
2819728
+5MB for the system libs.
Sure just 0.5MB when it is running...
I remember like it was yesterday that I complained vehemently that the text editor used up more than 17k.
"We are all bloated now"
Re: (Score:2)
Could be worse, could be emacs.
Re: (Score:2)
"eight megs and constantly swapping" used to be what emacs stood for. Imagine how svelte it would be if it were only eight megs today! Typed in a 173 meg firefox process, which is indeed positively svelte compared to the 700+MB it was yesterday - "firefox 4 is better and faster" my butt.
Re: (Score:2)
Firefox regularly uses more memory than a whole windows/ubuntu virtual machine
Re: (Score:2)
Re: (Score:2)
Good luck running dos 6.22 (that was the OS, not the 3.1 gui) on a modern workstation.
Re: (Score:2)
Holy reality check batman. I didn't say Win 3.11 would run at all let alone flawlessly. DOS 6.22 will NOT run at all, it won't even boot on anything newer than a P2. Also, the drivers would be for DOS 6.22 not windows 3.11 because windows 3.11 is a GUI not an operating system, it doesn't have drivers!
For extra fun you can try, and fail, to run Win95 (aka DOS 7) on a new system and crash and burn at the incompatible instruction set!
Re: (Score:2, Funny)
RSS is a news aggregator. /. is a news aggravator.
Re: (Score:3)
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: (Score:2)
Re: (Score:2)
Really? (Score:2)
First, at least the Flash I get on Ubuntu with Chrome is 32-bit. There was an x86_64 experiment for awhile, but have they actually kept it up, and is it actually stable? If so, I might install it manually, but I do like how my package manager keeps me up-to-date -- but it does so with 32-bit and nspluginwrapper.
Second, define "accelerated". It sure as hell isn't accelerated the way it is on Windows. Two major things seem to be lacking: First, my video card can do native H.264 decoding, but Flash doesn't app
Re: (Score:2)
My video card can do it as well, but Adobe hasn't decided to leverage VDPAU yet, this is not a linux problem. Adobe could also target VA-API, which would allow them to have a large amount of supported systems in one go.
Linux Flash playback is NOT hwaccelerated on Intel (Score:2)
I have the native 64 bit Flash plugin I can tell you that it is not THAT smooth and accelerated on many Intel GPUs. You may wish to qualify your statement with some links (the PenguinSWF blog [adobe.com] talks about the issues). On the same machine that struggles with certain Youtube Flash videos on Linux, the performance is fine under Windows. I believe you will need your GPU to have at certain GL features for support for significant Flash acceleration under Linux [adobe.com] and if your GL driver returns SGI string it will typic
Re: (Score:2)
Re: (Score:2)
Playing back 480p video taking up ALL the system resources on a machine that was probably sold within the last two years is, "Pretty good"?!?!
Re: (Score:2, Insightful)
Re: (Score:2)
To be fair, this is mainly the problem of nvidia and ati/amd not doing a proper job with linux drivers, rather then linux.
Then again, that doesn't really matter for end-user. He's the one that SOL.
Re: (Score:3)
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 t
Re: (Score:2)
Because in less than a year a new version of whatever distro i put on will come out and they WILL shit themselves and die if they are updated.
You keep insisting that you must be on the bleeding edge without the bleeding. If you're selling something to a regular not-so-savvy customer IT SHOULD NOT BE UPDATING EVERY SIX MONTHS. Until you get that through your thick skull and pick a sensible distro for your use, you're going to have problems and they're due to the person in your mirror. This is starting to sound like the joke with the patient that comes into the doctor's office saying "It hurts when I poke my eye" and the doctor says "Well, stop pok
Re: (Score:3)
Re: (Score:2)
When was your experience? I haven't had Flash problems in Linux for about 3 years. Maybe it's my hardware... but you shouldn't expect Flash to run smoothly on your hardware if it wouldn't run Windows but at a snails pace.
Re: (Score:2)
english is one of the most complicated languages, not by design but by exceptions to the rule
you can not even go 1 hour in any direction without hearing a totally different dialect
so keep assuming you know everything about written language there duck butter