×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Linux 2.6.34 Released

kdawson posted more than 2 years ago | from the fresh-bits dept.

Open Source 268

diegocg writes "Linux 2.6.34 has been released. This version adds two new filesystem, the distributed filesystem Ceph and LogFS, a filesystem for flash devices. Other features are a driver for almost-native KVM network performance, the VMware balloon driver, the 'kprobes jump' optimization for dynamic probes, new perf features (the 'perf lock' tool, cross-platform analysis support), several Btrfs improvements, RCU lockdep, Generalized TTL Security Mechanism (RFC 5082) and private VLAN proxy arp (RFC 3069) support, asynchronous suspend/resume, several new drivers and many other small improvements. See the full changelog here."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

268 comments

Penguin (-1)

Anonymous Coward | more than 2 years ago | (#32234802)

I fucked your dead great grandmother!! And a penguin!!1

a filesystem for flash devices (5, Funny)

deathcow (455995) | more than 2 years ago | (#32234844)

> a filesystem for flash devices

here we go again, unless we stop supporting flash, Apple has refused to distribute dual-boot Linux enabled iPads

Re:a filesystem for flash devices (0, Troll)

Hal_Porter (817932) | more than 2 years ago | (#32234858)

Here we go again, unless we stop supporting the Spanish Republicans Herr Hitler has threatened to invade Poland.

Re:a filesystem for flash devices (2, Funny)

hellop2 (1271166) | more than 2 years ago | (#32234876)

Is that a joke?

Re:a filesystem for flash devices (4, Funny)

Anonymous Coward | more than 2 years ago | (#32234892)

No. Flash support is in Linux and Apple indeed refuses to distribute dual-boot Linux enabled iPads.

Re:a filesystem for flash devices (5, Funny)

jo_ham (604554) | more than 2 years ago | (#32235028)

I'm more concerned that this is a slashvertisment for Linux.

Slashdot used to be about news for nerds, stuff that mattered!

How far it has fallen.

Re:a filesystem for flash devices (0, Offtopic)

WrongSizeGlass (838941) | more than 2 years ago | (#32235126)

If Linux isn't for nerds then who is it for? House fraus whose TV's are on the blink? Unrelenting overlords of evil corporate empires? Has-been movie starts and their hippie dippie 60 something rockstar friends? Well, maybe all of them, but especially nerds.

Re:a filesystem for flash devices (5, Funny)

Anonymous Coward | more than 2 years ago | (#32235242)

One of the new features of 2.6.34 kernel is Woooooosh owerflow protection

Re:a filesystem for flash devices (1, Insightful)

Anonymous Coward | more than 2 years ago | (#32235246)

If Linux isn't for nerds then who is it for? House fraus whose TV's are on the blink? Unrelenting overlords of evil corporate empires? Has-been movie starts and their hippie dippie 60 something rockstar friends? Well, maybe all of them, but especially nerds.

Woooosh

(that was the sound of the joke flying over your head)

Re:a filesystem for flash devices (0, Offtopic)

WrongSizeGlass (838941) | more than 2 years ago | (#32235732)

Whoosh indeed.

OK, so I'm a dumbass ... I claim "no coffee" and "I just had to scoop the cat box" as reasons my brain wasn't functioning on a full 3 1/2 cylinders this morning.

Re:a filesystem for flash devices (-1, Troll)

VTI9600 (1143169) | more than 2 years ago | (#32235188)

Yeah, I mean...its like, "News for Linux, Stuff that Linuxes" WTF?

Re:a filesystem for flash devices (3, Funny)

Rogerborg (306625) | more than 2 years ago | (#32235404)

Well, it's a kdawson post - he's notorious for posting Slashvertisements for the latest brand of rip off Snake Oil. I mean, "Linux", come on, what's that supposed to be? Some shoddy knock off of BSD/MacOS Unix? I bet it's just the periscope of a Microsoft patent submarine.

Re:a filesystem for flash devices (0, Troll)

Jesus_666 (702802) | more than 2 years ago | (#32235510)

Actually, Linux is a German detergent brand. I have no idea why we should care about detergent all of a sudden.

Re:a filesystem for flash devices (0)

Anonymous Coward | more than 2 years ago | (#32235730)

I am pretty sure they are talking about Flash memory devices...like a flash drive and NOT Adobe Flash...are people really this easily confused?

KVM (5, Interesting)

VTI9600 (1143169) | more than 2 years ago | (#32234862)

Other features are a driver for almost-native KVM network performance

KVM is fantastic virtualization technology, yet Xen gets all the hype these days. Why? Paravirtualization is pretty cool stuff, but seriously, what CPU's are made without some type of hardware-assisted virtualization support?

Re:KVM (4, Informative)

IBBoard (1128019) | more than 2 years ago | (#32234938)

Erm, quite a lot? Intel use it as one of their distinguishing factors between upper and lower tier chips (albeit one that they put in data sheets but don't make overly obvious).

Re:KVM (5, Informative)

VTI9600 (1143169) | more than 2 years ago | (#32235042)

I'm sure you are disappointed that your 200Mhz Pentium Pro doesn't support vt-x, but the rest of the world owns (or will soon purchase) processors that do. To see what I mean, just go to newegg.com. 63 out of the 76 (83%) desktop-class [newegg.com] processors they sell have virtualization technology built in. 78 out of the 80 (98%) of the server-class [newegg.com] (ones that really matter) processors they sell support it.

And, if you still don't believe me, check out this page [wikipedia.org] on Wikipedia for a list of the Intel processors that support VT-X. Among the crapload of processors listed, you'll notice that 100% of their newest, i3, i5 and i7 processors have virtualization support.

Re:KVM (2, Informative)

Calinous (985536) | more than 2 years ago | (#32235252)

If you want an inexpensive chip, you should carefully check Intel's support for virtualization - by example, some of the E7400 and E7500 had it, some didn't. Same for E5400 and E5300 (some have it, some don't).

Re:KVM (5, Informative)

icebraining (1313345) | more than 2 years ago | (#32235262)

Yeah, but then you have dick moves like my CPU supporting VT (AMD Neo) but being disabled by the Bios with no option to enable it. Thanks HP!

Re:KVM (1)

alen (225700) | more than 2 years ago | (#32235744)

my athlon 64 3200 had VT. no need to enable it in the BIOS like with intel either. it just works

Re:KVM (1)

L4t3r4lu5 (1216702) | more than 2 years ago | (#32236008)

I get the impression that the issue with the parent is regarding the motherboard, not the processor. HP have a habit of having their own bespoke boards in their desktop machines, more often than not with custom BIOS. Unless you have an HP computer (or Dell, Acer, $vendor machine) your situation will more than likely differ to theirs.

Re:KVM (1, Informative)

Anonymous Coward | more than 2 years ago | (#32235320)

*sigh* To everyone complaining about the shitty processors they got stuck with, I just want to say that for the past three (at least) years, I've been especially careful to make sure that the ones I bought had VT support. If you didn't do the same then maybe you didn't really need it that much to begin with.

Re:KVM (1)

jsolan (1014825) | more than 2 years ago | (#32235900)

Why would I want to spend money on buying new servers when the ones I own already work fine? Just because the next servers I buy will probably have vt-x, doesn't change the fact that the ones I currently own don't.

Using the argument that people just need to upgrade their hardware is garbage. Virtualization is supposed to CUT costs, not incur new hardware costs. The fact that I could take old machines that have run their course as database servers and convert them into several virtual machines for more lightweight stuff (web,dns,ldap,etc.) is what drew me into virtualization in the first place.

Re:KVM (1)

Klinky (636952) | more than 2 years ago | (#32235082)

Even lower-end chips are coming out with VM tech. I am running a Celeron E3200 @ 3.8ghz w/ VTd. Though my old Pentium-Dual Core E2160 did not support VTd. There is old hardware out there that virtualization is used on such as P4s and earlier Core/Core2 CPUs.

Re:KVM (4, Informative)

1s44c (552956) | more than 2 years ago | (#32235006)

Other features are a driver for almost-native KVM network performance

KVM is fantastic virtualization technology, yet Xen gets all the hype these days. Why? Paravirtualization is pretty cool stuff, but seriously, what CPU's are made without some type of hardware-assisted virtualization support?

Xen doesn't get all the hype. From what I've seen everyone is ditching xen and redhat is leading the way. Not that I mean to imply that xen deserves to get ditched, it's great too.

Re:KVM (2, Interesting)

arth1 (260657) | more than 2 years ago | (#32235428)

I think Redhat's dropping of Xen for KVM is as much politics as anything else. In the eyes of business, Xen = Citrix, and if going for Xen, why not go for Citrix?

Personally, I'm very pleased with Xen except for the qemu IO performance. Setting the host's block device schedulers to noop (for linux guests) or deadline (for Windows guests) helps, but high host IO load still makes it very hard to do advertised features like instant failover using an NFS-hosted container.

Re:KVM (1)

mr_da3m0n (887821) | more than 2 years ago | (#32235616)

Personally, I'm very pleased with Xen except for the qemu IO performance. Setting the host's block device schedulers to noop (for linux guests) or deadline (for Windows guests) helps, but high host IO load still makes it very hard to do advertised features like instant failover using an NFS-hosted container.

The solution to this would be to use PV drivers for HVM DomUs. This effectively closes the gap in performance between Paravirtualized DomUs and Fully Virtualized DomUs. Commercial XenServer provides them, and people routinely Build them. They are a bit of a pain to install on vanilla Xen DomUs, for they are not signed, and require a boot argument to be added to windows (/GPLPV) but they work as advertised.

Re:KVM (0, Redundant)

Jurily (900488) | more than 2 years ago | (#32235072)

but seriously, what CPU's are made without some type of hardware-assisted virtualization support?

Umm.. most of them?

Re:KVM (1)

mr_da3m0n (887821) | more than 2 years ago | (#32235632)

I would like to return your implicit question: What does KVM has that Xen has not? Why is everyone suddendly switching to it?

Re:KVM (1)

shallot (172865) | more than 2 years ago | (#32235748)

Other features are a driver for almost-native KVM network performance

KVM is fantastic virtualization technology, yet Xen gets all the hype these days. Why? Paravirtualization is pretty cool stuff, but seriously, what CPU's are made without some type of hardware-assisted virtualization support?

Er, it's KVM that gets all the hype these days because it's still got some novelty. Xen just has the users, because it's simply more mature.

All Very Nice But... (4, Insightful)

mrpacmanjel (38218) | more than 2 years ago | (#32234866)

Is the RT2500-based chipset working reliably now?

The developers switched to a new driver model because it's "better".

If "better" means once-working wifi chipset becomes grossly unstable, previous drivers are considered "legacy" hence will not compile on kernels later that 2.6.29 and current drivers are as stable as a "one-legged man playing football".

A few years later and 2.6.34 is released - is it working yet?

Considering the RT2500 chipset is present many wifi products the current state of "stability" is woefully inadequate.

(and don't get me started on f***ed up i845 drivers for xorg! - worked fine under previous kernels & xorg an update later by both - graphics performance royally screwed and many crashes)

Apart from that - happy Linux user for over 10 years!

Re:All Very Nice But... (4, Insightful)

dbIII (701233) | more than 2 years ago | (#32234920)

Is the RT2500-based chipset working reliably now?

Here's how the dismal state of support for that chipset was explained to me.
The answer is probably that mine has worked for years and yours hasn't. The really annoying thing is a lot of slightly different things have come out under that name and even under MS Windows if you don't use the driver that came with it you are stuffed - a driver for another undocumented variant won't help.

Re:All Very Nice But... (1)

Chrisq (894406) | more than 2 years ago | (#32234922)

Though the RT2500 support has been slow in coming it has been stable for over a year now.

Re:All Very Nice But... (4, Informative)

HateBreeder (656491) | more than 2 years ago | (#32234982)

The problem is that the RT2500 chipset is proprietary, closed-source that's "maintained" by a Taiwanese manufacturer who doesn't care about his users at all and only wants to sell cheap hardware and as much of it as possible.

Why would you get quality, polished drivers that are updated to support newer paradigms in newer kernels if the manufacturer isn't cooperating?

I think it's magic that these drivers work at all.

Next time, buy better kit with a reputable mfr that cares about linux support.

Re:All Very Nice But... (4, Insightful)

mrpacmanjel (38218) | more than 2 years ago | (#32235136)

The problem for me is that the "legacy" drivers were rock solid and I never thought about it until kernel 2.6.30 & greater were released.

My wifi was ultra-reliable under the "legacy" drivers.

Since the newer drivers were released I have had nothing but problems.

What changed between old and new drivers?

Re:All Very Nice But... (1)

HateBreeder (656491) | more than 2 years ago | (#32235170)

Would you use windows 98 drivers with windows 7?

They most likely would not work.

You would download newer versions of the drivers... right? you wouldn't complain to microsoft that your win98 drivers aren't working anymore, would you?

So why do you think the same situation should just magically work in Linux?

Re:All Very Nice But... (0, Troll)

mrpacmanjel (38218) | more than 2 years ago | (#32235190)

Applying that argument to Linux is retarded.

Re:All Very Nice But... (0)

Anonymous Coward | more than 2 years ago | (#32235250)

This is the problem with Linux as opposed to a proprietary environment like Windows or OS X.

In proprietary land, backwards compatibility is pretty much mandatory, supporting older existing software is paramount even if it means compromising on elegance or "cleanliness" of design. In Open land, the maintainer can rewrite their subsystem on a whim without even bothering to create a one or two year roadmap for people to prepare.

Talk all you want about "technical superiority", stuff that works is generally always better than stuff that doesn't.

[Linux user since 2001]

Re:All Very Nice But... (2, Interesting)

AvitarX (172628) | more than 2 years ago | (#32235206)

Because as a Linux user I have become quite accustomed to things "just working".

Unfortunately, wireless destroys that notion, causing irritation (though my thinkpad T500 shows me how it is when the stuff does work, and it's great).

Re:All Very Nice But... (2, Informative)

mrpacmanjel (38218) | more than 2 years ago | (#32235256)

Generally Linux works very well.

For me the two biggest problems seems to be wifi & graphics cards.

ATI decided my r300-based card was legacy and discontinued it via the closed-sources drivers. I'm screwed(thankfully the open source drivers are ok but nowhere near as fast).

RT2500 - I could download the source of the serialmonkey drivers and compile them. Great it works fine and did that with every distro upgrade.

Then these drivers were abandoned and all focus is now on the in-kernel version and stability has suffered ever since.

I would have thought the maintainers could have adapted the legacy driver to work with the new kernel - even as a temporary solution.

Then again Linux is a "server" OS and seen that way from the kernel maintainers.

Re:All Very Nice But... (1)

e70838 (976799) | more than 2 years ago | (#32235340)

I switch to Linux exactly because windows vista64 and windows 7 do not provide drivers for my old printer and my old scanner. No driver problem with Linux, because Linux supports legacy hardware.

Re:All Very Nice But... (3, Informative)

stsp (979375) | more than 2 years ago | (#32235708)

The problem is that the RT2500 chipset is proprietary, closed-source that's "maintained" by a Taiwanese manufacturer who doesn't care about his users at all and only wants to sell cheap hardware and as much of it as possible.

Well, actually, Ralink has for a long time been providing documentation to open source developers writing drivers for their devices, without requiring an NDA.

Re: RT2500 support (0, Troll)

kv9 (697238) | more than 2 years ago | (#32235562)

if you want stable and good wireless support you should use OpenBSD (go ahead linuxfags, mod me down). HTH.

Excellent (5, Funny)

Anonymous Coward | more than 2 years ago | (#32234884)

With releases like these, it's no wonder M$ is getting worried. Been running this kernel a while now on our production servers (even from before it it was tagged release, I like running bleeding edge in order to get the most performance from my company's hardware investment) and save from a few data corruptions issues, it's been rock stable! I have to play with the new KVM support later on one of the servers with the least amount of customers on it (couple of hundreds), looks nice!

Sadly... it looks like my company is looking at going with Windoze for a few important servers because of a few outtages. I know it was because of faulty hardware, because I had just compiled a custom kernel for those servers with just the right flags needed (I want to get the most performance!) but this must have triggered a hardware bug because the kernel worked fine on my work laptop. Sigh...

Anyway, keep up the good work!

Re:Excellent (1, Insightful)

putaro (235078) | more than 2 years ago | (#32234910)

Hmmm...sounds like your company would be better off hiring a new systems administrator than going with Windows. Good thing you're posting as AC!

Re:Excellent (-1, Redundant)

Anonymous Coward | more than 2 years ago | (#32234992)

Yeah... that's not going to happen. Not only am I good at what I do (seriously, how many system administrators even take their time to compile custom kernels nowadays for maximum performance?), but a very close relative is also the CEO. I bet people here even use us for their hosting :)

Re:Excellent (2, Insightful)

cduffy (652) | more than 2 years ago | (#32235020)

seriously, how many system administrators even take their time to compile custom kernels nowadays for maximum performance?

That's a bug, not a feature.

"Maximum performance"? If it's storing customer data, I want a kernel that's had heavy automated testing, a full round of manual QA, and validation by my hardware vendor, thank you very much.

Re:Excellent (-1, Troll)

Anonymous Coward | more than 2 years ago | (#32235076)

but a very close relative is also the CEO.

Must be a Gook, Stinky(Indian), or Sandnigger.

Nope, because you can actually speak English. Are you gonna reimburse us for our downtime, Mister inbred fuck from Georgia?

Re:Excellent (0, Offtopic)

nacturation (646836) | more than 2 years ago | (#32235102)

Not only am I good at what I do (seriously, how many system administrators even take their time to compile custom kernels nowadays for maximum performance?), but a very close relative is also the CEO. I bet people here even use us for their hosting :)

Hopefully not, as a competent administrator knows that data integrity is far more important that a fractional improvement in performance.

but this must have triggered a hardware bug because the kernel worked fine on my work laptop

Exactly. A competent admin would have tested the config on identical hardware as what is running in production. That you consider a laptop to be sufficient testing demonstrates either a lack of funding for your "the CEO and I share DNA" company or that you suffer from a lack of training.

Sadly... it looks like my company is looking at going with Windoze for a few important servers because of a few outtages.

Had you properly designed your system for redundancy, you wouldn't have experienced those issues. Or if you weren't dicking around with kernels that should not be used in production, you similarly would not have experienced the outages. I take it you're going to be running the latest Windows Server alpha and beta releases from Microsoft in order to get maximum performance?

Re:Excellent (1, Funny)

Anonymous Coward | more than 2 years ago | (#32235146)

Successful troll is successful, I guess. Well done.

Re:Excellent (0)

Anonymous Coward | more than 2 years ago | (#32235192)

Exactly. A competent admin would have tested the config on identical hardware as what is running in production. That you consider a laptop to be sufficient testing demonstrates either a lack of funding for your "the CEO and I share DNA" company or that you suffer from a lack of training.

For the record, obviously I was testing the kernel in a virtual machine on my laptop, using the same version of the distribution we use (in-house, built by me, but originally based on a now ancient version of Red hat).

Had you properly designed your system for redundancy, you wouldn't have experienced those issues. Or if you weren't dicking around with kernels that should not be used in production, you similarly would not have experienced the outages. I take it you're going to be running the latest Windows Server alpha and beta releases from Microsoft in order to get maximum performance?

Either a customer's data is on a server, or it is not. If that machine fails, it doesn't matter if we are running Linux or Windoze or whatever, you still have to startup a spare machine and restore from backups. I just know that if we were running Windoze, we (I) would have to restore failing machines far more often than now, that's all.

Re:Excellent (1)

1s44c (552956) | more than 2 years ago | (#32235464)

For the record, obviously I was testing the kernel in a virtual machine on my laptop, using the same version of the distribution we use (in-house, built by me, but originally based on a now ancient version of Red hat).

That's sick, ugly, and wrong. What's worse is you actually think you are doing the right thing.

Stop messing about and replace all the custom stuff with standard stuff before you drag your company to its knees or drive it into the arms of Microsoft.

Re:Excellent (2, Interesting)

linuxgurugamer (917289) | more than 2 years ago | (#32235350)

I used to compile kernels, but stopped. There is no way that I, as a single administrator, can perform all the necessary testing to assure that there aren't any kernel problems. So now I don't, but now I get 100% uptime instead. I prefer the uptime to the performance.

Re:Excellent (3, Insightful)

1s44c (552956) | more than 2 years ago | (#32235400)

Yeah... that's not going to happen. Not only am I good at what I do (seriously, how many system administrators even take their time to compile custom kernels nowadays for maximum performance?), but a very close relative is also the CEO. I bet people here even use us for their hosting :)

There is a massive amount of difference between being able to compile a custom kernel and being in the kind of situation where it's the right thing to do. 'Good at what you do' doesn't mean technically brilliant, it means doing the right thing at the right time.

Keeping your job because you are related to the CEO is the kind of nepotism that kills otherwise good companies.

Whooosh (1, Informative)

Anonymous Coward | more than 2 years ago | (#32235482)

Is it a bird? Is it a plane? Something just flew way overhead....

Re:Excellent (1)

westlake (615356) | more than 2 years ago | (#32234974)

It looks like my company is looking at going with Windoze for a few important servers because of a few outtages. I know it was because of faulty hardware, because I had just compiled a custom kernel for those servers with just the right flags needed (I want to get the most performance!) but this must have triggered a hardware bug because the kernel worked fine on my work laptop...

Help Wanted, Male.

Re:Excellent (3, Insightful)

1s44c (552956) | more than 2 years ago | (#32235034)

You don't put bleeding edge or custom kernels on production servers without seriously heavy testing. You would not run production stuff on a windows beta release would you? It's the same thing.

Stick to proper releases of good distributions and customize as little as possible. You will get a system many times more stable than anything MS has ever come up with.

Re:Excellent (1)

solanum (80810) | more than 2 years ago | (#32235490)

Whooooossssshhhhhh!!!! That was the grand-parent post going over your head.

Now why was the parent marked insightful exactly?

Production Bleeding Edge == 2.6.32 (0)

Anonymous Coward | more than 2 years ago | (#32235284)

2.6.32 is what you should consider bleeding edge. 2.6.27 would be safe.

Re:Excellent (2, Insightful)

shish (588640) | more than 2 years ago | (#32235304)

Why are all the replies to this comment seeming to take it seriously? :-|

Re:Excellent (2, Informative)

linuxgurugamer (917289) | more than 2 years ago | (#32235360)

Because I've heard the exact same thing from people who actually believe it and have done it at their job. It is a comment made by a young, inexperienced person (I can't call them an administrator) who doesn't have the experience to understand the problems with doing this.

Re:Excellent (2, Insightful)

Kjella (173770) | more than 2 years ago | (#32235476)

Because I've heard the exact same thing from people who actually believe it and have done it at their job. It is a comment made by a young, inexperienced person (I can't call them an administrator) who doesn't have the experience to understand the problems with doing this.

What, can't be. According to slashdot all Linux administrators are born as black belt Linux experts and Windows administrators are all people that got lucky bumbling through their MSCE exam. Usually in comparison where five incompetent Windows administrators could be replaced with one competent Linux administrator, even though you could probably replace five incompetents with one competent one in general.

Re:Excellent (3, Insightful)

1s44c (552956) | more than 2 years ago | (#32235538)

What, can't be. According to slashdot all Linux administrators are born as black belt Linux experts and Windows administrators are all people that got lucky bumbling through their MSCE exam.

Noone is born a black belt at anything. You have to work at it. There are inexperienced Linux admins just like there are inexperienced windows admins. The ones who can't or don't want to learn end up on windows eventually.

Re:Excellent (2, Informative)

arth1 (260657) | more than 2 years ago | (#32235460)

Why are all the replies to this comment seeming to take it seriously? :-|

Because (a) it is Monday morning, and (b) Sturgeon's law applies to /. posters too.
And, unfortunately, (c) there are idiots like that out there. But they generally don't change their posting prefs to AC when bragging about their latest folly...

Re:Excellent (1)

1s44c (552956) | more than 2 years ago | (#32235550)

(c) there are idiots like that out there. But they generally don't change their posting prefs to AC when bragging about their latest folly...

The real idiots never figured out how to log in.

Re:Excellent (1)

1s44c (552956) | more than 2 years ago | (#32235512)

Why are all the replies to this comment seeming to take it seriously? :-|

Because some of us have had to clean up the mess left by people like him. The world is full of people who do really stupid things whilst thinking they are doing a good job.

Re:Excellent (2, Informative)

linuxgurugamer (917289) | more than 2 years ago | (#32235312)

Huh? You are running unreleased kernels, you _admit_ that you have "data corruptions issues" and you claim "rock stable"?

What idiot runs beta kernels on production servers? I'm glad you aren't working for me, because I'd fire your ass for doing this.

Production servers are NOT the place to run beta kernels.

And you are complaining because your company is going with Windows "because of a few outages"? How do you know that it wasn't a kernel bug triggered by that hardware configuration? Your laptop has different hardware than your servers, you simply cannot assume that since it runs fine on your laptop that it will also be fine on the server.

People like you annoy me. "Most performance" does NOT equal good business. The most important thing to a business is total reliability. Play with the new stuff on a test system, not on a production system

I'm not impressed (1, Funny)

justleavealonemmmkay (1207142) | more than 2 years ago | (#32234888)

Windows reached version 3.0 some 20 years ago...

Re:I'm not impressed (0)

dropadrop (1057046) | more than 2 years ago | (#32234926)

I agree, a major change is needed to make Linux more convincing. I vote the next kernel be numbered 10.2.35. This should prove to everyone that Linux is mature, and ready for the desktop.

Re:I'm not impressed (1)

K. S. Kyosuke (729550) | more than 2 years ago | (#32235486)

Windows reached version 3.0 some 20 years ago...

My operating system has reached stable version 23.2. Take that!

GPU switching (3, Insightful)

TheLink (130905) | more than 2 years ago | (#32234914)

> Some laptops have two GPUs, a low-power and inefficient GPU and a high-power and powerful GPU. Users should be able to switch to one or another at runtime. In this version, Linux adds support for this feature. You need to restart X, though.

How do you restart X without affecting all your GUI apps? If you can't restart X without bringing down your GUI apps, I don't see the point for the target audience.

For some people, "only having to restart X" will only save a bit of time over rebooting the whole laptop, reconfiguring bios etc.

Re:GPU switching (3, Informative)

TeknoHog (164938) | more than 2 years ago | (#32234942)

How do you restart X without affecting all your GUI apps? If you can't restart X without bringing down your GUI apps, I don't see the point for the target audience.

If you are using something like Gnome or KDE, it can probably save your GUI session. Individual applications will have to deal with their contents, but many of them already do that. At least Firefox and Openoffice can restore their sessions after being terminated.

Re:GPU switching (2, Informative)

FeepingCreature (1132265) | more than 2 years ago | (#32235386)

If you are using something like Gnome or KDE, it can probably save your GUI session. Individual applications will have to deal with their contents, but many of them already do that. At least Firefox and Openoffice can restore their sessions after being terminated.

In KDE, System Settings -> Advanced -> Session Manager -> On Login, Restore Manually Saved Session. After that, you can save your session state from the logout menu or, alternatively, using a shellscript that loops every 30s or so and does

# KDE3
dcop ksmserver ksmserver saveCurrentSession
# or KDE4
dbus-send --dest=org.kde.ksmserver /KSMServer org.kde.KSMServerInterface.saveCurrentSession

Re:GPU switching (1, Interesting)

Anonymous Coward | more than 2 years ago | (#32234968)

Well, AFAIK, until veeery recently (Still beta I think) you basically have to do the same on windows. Close all apps and restart the desktop.

Re:GPU switching (1, Informative)

Anonymous Coward | more than 2 years ago | (#32235258)

Well, AFAIK, until veeery recently (Still beta I think) you basically have to do the same on windows. Close all apps and restart the desktop.

Don't know about other configs, but on laptop with ATI graphics and Windows 7 the switch between discrete and integrated graphics is seamless. No need to close anything down or restart anything, open windows stay open. It can also be set on auto, controlled by battery level. And it has been working like this at least since Windows 7 launched last October.

Re:GPU switching (2, Informative)

skynexus (778600) | more than 2 years ago | (#32235016)

If you can't restart X without bringing down your GUI apps, I don't see the point for the target audience.

For some people, "only having to restart X" will only save a bit of time over rebooting the whole laptop, reconfiguring bios etc.

Not all laptops have a BIOS configuration that allows you to choose the GPU (ASUS UL series for instance). On mine, I had to change the SATA operation mode to have the second GPU work, but this in turn meant a severe performance degradation on my SSD. Without that (deficient) improvisation, I would not have been able to use the second GPU at all!

Besides, logging out of your desktop and then logging in again is surely better than what you suggest?

Re:GPU switching (3, Insightful)

Kjella (173770) | more than 2 years ago | (#32235088)

Good question, but wrong project. The kernel is only responsible for initializing, suspending, resuming and lately modesetting of the hardware and it seems that is possible now. There probably needs to be some userspace code to pull information from one GPU and load it into the other but that's for the xorg server to do. They're probably working on it but it won't be in a Linux (the kernel) release announcement.

Re:GPU switching (1)

ettlz (639203) | more than 2 years ago | (#32235198)

The kernel is only responsible for initializing, suspending, resuming and lately modesetting of the hardware and it seems that is possible now.

Plus managing GPU memory allocation. But yes, this is probably something to be added to XRandR, or some other protocol extension. (What would happen to normal, non-X virtual consoles, though? This might require some more stuff in the kernel.)

Re:GPU switching (1)

ledow (319597) | more than 2 years ago | (#32235294)

I'm not saying this is the solution, nor how it should be done, but you could conceivably run a "remote" X-Windows session on a virtual buffer on the laptop - connecting to it with another X-Windows client on the same machine - and then, when you "switch" GPU's, the restart of X-Windows will only affect the "client" viewing the real X-session but be transparent to the user because they'll reconnect to their original session.

It's not a huge stretch of the imagination that the virtual buffer can pass off necessary acceleration / state to the client, whatever that happens to be. However, it's more likely that in a month's time every X Server on the planet will start supporting some sort of underlying client refresh whenever it detects a switch and will be seamless. It's *JUST* made it into a kernel, for goodness sake, don't expect client support until the authors actually have something to program against.

Re:GPU switching (1)

icebraining (1313345) | more than 2 years ago | (#32235310)

I agree that's not as useful as it could be, if it was able to do it on the fly easily.

But it could theoretically (I've never tried it) be done using Xmove [wikipedia.org], which "allows the movement of X Window System applications between different displays and the persistence of X applications across X server restarts".

xmove lets the client disconnect from its current X server, and connect to a new one, at any time. The transition is completely transparent to the client. xmove works by acting as a proxy between the client and server. It is a "pseudoserver" which stores enough server state so that clients can connect to a new server without being disrupted.

Hurrah - now I'll get pre-built nVidia drivers! (1)

IBBoard (1128019) | more than 2 years ago | (#32234958)

I put an openSUSE Build Service version of the .34RC kernel on my new desktop because it fully supported the new Core i5 I'd just installed. Down-side was that there weren't any pre-built nVidia drivers because it wasn't a final kernel yet. Hopefully nVidia will start building the drivers in their repo so that I can move to a repo for my drivers :)

The most exciting thing is (5, Interesting)

Advocadus Diaboli (323784) | more than 2 years ago | (#32235044)

that this kernel already got device IDs for next years Intel hardware. This is something completely new, since Intel so far had a much more closed policy and wouldn't have told device IDs prior to the chipset release.

Now there is a really good chance that driver code will make it into the distribution kernels until the new hardware will be released for mass production. So the chances that brand new hardware will work without any flaws in 2011 are higher than ever before.

Thanks to Intel for this change in their policy. This was a small step for Intel (since everybody "knows" that they will release new chips every year) but a giant leap for providing Linux hardware compatibility right "out-of-the-box".

Re:The most exciting thing is (0)

Anonymous Coward | more than 2 years ago | (#32235508)

Sorry but when was the last time Linux could not detect your CPU? I think priority for these sort of things should be with wifi cards, gpus etcetera.

Trim (1)

djfake (977121) | more than 2 years ago | (#32235430)

I believe that 2.6.34 will natively support TRIM, which will keep most second generation SSDs running clean. http://en.wikipedia.org/wiki/TRIM/ [wikipedia.org]

Re:Trim (1)

arth1 (260657) | more than 2 years ago | (#32235502)

Native support for ATA discard (better known as TRIM in the Windows world) appeared in 2.6.27, unless I'm mistaken.
Support for individual file systems depends on how quickly the FS devs have been able to add their part, but some distros have had working discard with their default file systems for a while now. At far lower kernel versions than 2.6.34 :-)

In reading kernel changelogs... (5, Funny)

fuzzyfuzzyfungus (1223518) | more than 2 years ago | (#32235588)

I'm always amused by at least one strange juxtaposition of the big-serious-enterprise-server stuff that the corporate devs are most interested in and the oddball hobby projects that can get included as well, so long as they follow the kernel process.

In this case, I think it was all the "multi-petabyte scaleable filesystem, esoteric btrfs improvements, kernel virtualization networking stuff, gamecon: add rumble support for N64 pads" that did it.
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...