“Scalability enhancements, as well, will add to the appeal of the latest kernel, Version 2.6, for enterprise customers, according to Linus Torvalds, the creator and top programmer for the Linux kernel, in an e-mail exchange last week with eWeek. While he would like the 2.6 kernel to be ready early next year, Torvalds said, “It’s just too hard to predict, and it does end up depending a lot on how good the vendors are at trying to calm things down through stability fixing.”” Read the report at eWeek.
Linux 2.6 on Horizon
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
This 2.6 kernal only excites the mainframe crowd given the lack of comments on this. There’s nothing here for the desktop enthusiast who is probably 95% of the posters here on OS News.
I do think it’s right for Tovalds to go with 2.6 and not 3.0 for this. If finally code is frozen and released in 1Q 2003, it’ll be at least another 9 month for any distro to really implement this.
Vic
There are already a “feature freeze” so to speak on 2.5 since a couple of weeks back so from here it’s only bugfixes and in general most bugs are very very unoticable. There IS features in 2.5 that clearly benefit desktop users, just that many of them have been backported and the main issue for desktop users isn’t kernel speed as much as XFree speed.
I’ve been running 2.5 kernel since around 2.5.30 on and off and as far as I’m concerned it’s already good enough to as permanent replacement for my old 2.4.19 kernel.
Nope, the concrete feature freeze is 12 January. At that time absolutely NO features are going to be added. Up until the Linus will continue to accept additions as so long as it doesn’t stuff things up.
As for kernel speed. What happens in the kernel does affect the XFree86 speed. Crap threading and process handling equals crap XFree86 eXPerience. The DRM modules have been improved, DirectFB has bee improved also, so there will be a user improvement. The big improvement will be on multi-processor boxes. As for Mainframe Linux, the idea was to sell a mainframe with a working Linux version NOW, and then upgrade to 2.6 will include a tonne of enhanements that will boost the performance ot the IBM mainframe Linux.
I’m using 2.5.49 right now, and X feels much faster, though faster than a snail isn’t necessarily “fast”.
Let hope XFree 5.0 will be the improvement we need.
“This 2.6 kernal only excites the mainframe crowd given the lack of comments on this. There’s nothing here for the desktop enthusiast who is probably 95% of the posters here on OS News.”
Not completely true, they also improved the I/O performance, and this should be very good for the desktop (so it won’t shock less), it should be much more responsiveness! As loyal desktop user, I can’t wait
“Let hope XFree 5.0 will be the improvement we need.” Ohh sorry have to reply on this one too
Xfree is a very good/fast server/client protocol already. If it’s slow, you should blame the windowmakers for it, there is a massive difference between the difference windowmakers.
Maybe it’s more because this is freaking old news. The only newsworthy part of this whole thing is it’s in eweek. All the contents of the article came up 2 weeks ago.
Linux 2.6 will be very interesting from a desktop perspective. The advances in threading, the fully preemptive kernel and the improved (hopefully) VM will make Linux much more responsive. It might finally be able to do stuff BeOS was known for, e.g. processing audio data with very low latency.
Have you noticed that in order to have good audio playback in XMMS, you need to set the buffer to almost half a second? And when you turn up the volume, it takes half a second to have an effect? That will no longer be the case if the new Linux kernel is as good as people say it is.
The only thing I am really missing is some way to get change notifications when files are changed to implement some kind of live queries.
regards,
tuttle
(who has really high hopes for 2.6)
XFree is still the big bottleneck for desktop users, ofcourse Kernel improvements effect overall computing but you can boost and improve the kernel all you want, I will still be able to see stuff repaint on my desktop, there have to be something fundamentally wrong somewhere.
XFree function wise leaves little to desire, just wish it speed wise could compete with the other OS’s, but being functional is more important cause it allowed me to totally replace my old WindowsXP install, but I sometimes miss the instant responsive GUI….
As far as I can tell, XFree is slow mainly because it uses the TCP/IP stack, even for local machines.
The XFS filesystem with the ability to start having massive files should be useful for the desktop crowd. Video and music editing are some of the main apps which are still challenging on PCs and XFS should help.
Exactly how qualified are you to say something like that? X uses UNIX domain sockets for local communication, and experiments with shared memory communication found a negligible speed-up. The pipe between client and server just isn’t the bottleneck.
tuttle:
Or how about when changing a setting on your equalizer…. this is also a problem in Windows… it’s very irritating.