Marcelo has announced the availability of the 2.4.30 kernel; no changes were made after 2.4.30-rc4. Since 2.4 is in deep maintenance mode, there is little in the way of new features in this release. It does contain a number of security updates and other important fixes, though.
The 2.4 series … the last Linux kernel series that still maintained some sense of respectability and professionalism. 2.6 is a developer’s playground, and a complete joke as far as professional development attitude goes.
The 2.4 series … the last Linux kernel series that still maintained some sense of respectability and professionalism. 2.6 is a developer’s playground, and a complete joke as far as professional development attitude goes.
Really? Well lets look at the facts.The early 2.4 kernels where a complete mess. 2.6 has matured much more quickly.
You can always just stick with a BSD.
(I love the BSD’s btw
I totally agree with you. 2.4 series are actually faster kernels than 2.6 – the only thing 2.6 got going for it is better ACPI, USB, Sata support. 2.4 feels just a bit more snappier than 2.6 on the same system.
I totally agree with you. 2.4 series are actually faster kernels than 2.6 – the only thing 2.6 got going for it is better ACPI, USB, Sata support. 2.4 feels just a bit more snappier than 2.6 on the same system.
I’m a *BSD user so I don’t follow linux development that much, but… I thought linux 2.6 beat 2.4 in almost all benchmarks?
e.g. http://software.newsforge.com/article.pl?sid=04/12/27/1243207&from=…
If you want to improve system responsiveness for the 2.4 kernel series (low-latency + preemption), use the ‘lck’ patchset:
http://www.plumlocosoft.com/kernel/
I’m a *BSD user so I don’t follow linux development that much, but… I thought linux 2.6 beat 2.4 in almost all benchmarks?
e.g. http://software.newsforge.com/article.pl?sid=04/12/27/1243207&from=…
“I was surprised how well 2.4 had done, as I had somewhat expected 2.6 to show at least a noticeable, if slight, increase over 2.4. Instead, they took turns besting each other from test to test — and in scalability — for a fairly even overall showing.”
> e.g. http://software.newsforge.com/article.pl?sid=04/12/27/1243207&f…
Read the article:
Both Linux 2.4 and 2.6 had the strongest showing overall for these tests, dominating just about every benchmark no matter the workload. Scalability for both kernels was also excellent with addition of an extra processor. In fact, I was surprised how well 2.4 had done, as I had somewhat expected 2.6 to show at least a noticeable, if slight, increase over 2.4. Instead, they took turns besting each other from test to test — and in scalability — for a fairly even overall showing.
Most desktops and embedded systems are 1 cpu (+ HT). Even where 2.6 beats 2.4, it’s marginally better than 2.4 and not counting the tweaks like ingo’s 0/1 scheduler or kernel latency patches that can make 2.4 beat 2.6 on all but the > 4 processor tests. Howmany of you have > 4 processors on your desktops/workstations raise your hands…I’m seeing 1 or 2 hands at the back of the room
If you look at embedded systems (wifi, routers, etc), everybody’s still on Linux 2.4 because it’s got a stable performance heuristic – linux 2.6 can have performance swings. THe last good 2.6 version was Linux 2.6.7.
Hey guys – running linux in the enterprise and still running with kernel 2.4.x. can’t justify the changes in 2.6 to upgrade
Who is still using the old 2.4 kernel? 2.6 is stable like a rock and standard.
> Who is still using the old 2.4 kernel?
Try TiVo.
As Milione pointed out its all up to what you need. Most people in enterprise are still running 2.4 just because they dont see the need to up to 2.6.
2.6 as mentioned is just a for introduction of new features 2.4 is more of a stable/secure kernel.
KDE
A kernel developer himself was quoted as saying “We don’t know which 2.6 kernels are stable and which aren’t — you just have to try them and see”. WTF? I paraphrased, BTW, but the message was the same. Hello … here I was assuming that the 2.6 branch means a stable branch. 2.5 was a time to play. Anything released as 2.6.x should be stable as rock. What a joke.
Anyway, needless to say, I run real operating systems: FreeBSD on the servers, and XP SP2 on the desktops.
real for you maby if you cant find a stabel 2.6 kernel its your ignorance. for me its rock solid.
the last time i tried freeBSD on my hw i ran xf86cfg and that mad te computer crasch then i wanted to reboot and the kernel image was corrupt so it couldnt boot.
then there is two options
1 freeBSD is crappy
2 i dont know freeBSD and did somthing wrong
i think 2 is the rigth one and i think thats why you cant get linux stable
Firstly you can´t even sync Palm on 2.6
I´ve heard rumours a few aliens managed but briefly to make that work.
In an internet café Knoppix connectivity works when 2.4 is running but not 2.6? why?
I mean I am all for the continuous development of later kernels but 2.4 does seem very stable probably better.
2.6 has the features
2.4 has the real stability
2.6 development has been a mess. We’re seeing poorer performance now at 2.6.11 than on the original 2.6.0. Lots of bug fixes though 😉
When I upgraded to a 2.6 kernel (about six months after the release) I noticed NO improvement in speed, despite the so called “pre-emptive” patches etc. Load of codswallop if you ask me.
The only real advantage of the 2.6 kernel series is the support for newer stuff. That’s it. It loses out on stability etc.
Let’s see what’s wrong (as I perceive it)…issues with burning CDs as a non root user. Udev is well…poorly documented with very few, if no tools/utilities for it. Constantly in flux, no stability.
Bonuses? Inclusion of alsa, better i2c support, inclusion of SELinux, most of which can and should be backported to the 2.4 series providing it doesn’t destabilise it. Nicer and easier to build and compile. Better hardware support.
The kernel developers have lost the plot and sold out to the big businesses here I suspect. We have kernel developers telling us that “don’t expect stable kernels from kernel.org, it’s up to the distros to make the kernels stable”. I’m like “what the fuck”? How are the smaller distros going to do this? This new kernel development process? Keeping 2.6 in constant flux, non development of a parallel 2.7 tree (something which has server the Linux kernel development process very well for quite some time I might add), this sub patch monstrosity, etc?
To be honest, 2.6 kernels are not worth the extra effort. 2.4 does most of what you need to do, the only issue is with hardware support. I see no real advantage to using a 2.6 kernel, as do many others. Real, sticky issues need to be solved – like a ABI within the kernel, that’s standard, get rid of the linking libraries approach. Until this fundamental thing is accomplished you will see a lack of 3rd party software developers porting their software to run on Linux. That will mean less market share and less users.
Dave
”
Let’s see what’s wrong (as I perceive it)…issues with burning CDs as a non root user. Udev is well…poorly documented with very few, if no tools/utilities for it. Constantly in flux, no stability. ”
issues with burning cds was a single version regression long since fixed. udev is very well documented. just google for it and despite frequent updates is HIGHLY stable and is being used by nearly all modern distributions. so
I really do love when it comes to Linux kernel all those people that claim to perfectly know all the errors, mistakes and whatever done by the kernel team, while none of them have never contributed even a single two-lines patch. It’s funny. It’s OSNews forum.
2.4 sucked much more in 2.4.x early days.
It’s funny how people talk about 2.6 kernel. Looks like there’s someone putting a gun in their heads, forcing them to use 2.6. Come on, stay with 2.4 if you think 2.6 is a “playground”.