Linux Kernel 4.6 Officially Released (softpedia.com)
An anonymous coward writes: Just like clockwork, the Linux 4.6 kernel was officially released today. Details on the kernel changes for Linux 4.6 can be found via Phoronix and KernelNewbies.org. NVIDIA GeForce GTX 900 Maxwell support and Dell XPS 13 Skylake support are among the many hardware changes for 4.6. For Linux 4.7 there are already several new features to look forward to from new DRM display drivers to a new CPU scaling governor expected.
prisoninmate also writes: Linus Torvalds announced the final release of the anticipated Linux 4.6 kernel, which, after seven Release Candidate builds introduces features like "the OrangeFS distributed file system, support for the USB 3.1 SuperSpeed Plus (SSP) protocol, offering transfer speeds of up to 10Gbps, improvements to the reliability of the Out Of Memory task killer, as well as support for Intel Memory protection keys," [according to Softpedia].
"Moreover, Linux kernel 4.6 ships with Kernel Connection Multiplexor, a new component designed for accelerating application layer protocols, 802.1AE MAC-level encryption (MACsec) support, online inode checker for the OCFS2 file system, support for the BATMAN V protocol, and support for the pNFS SCSI layout."
prisoninmate also writes: Linus Torvalds announced the final release of the anticipated Linux 4.6 kernel, which, after seven Release Candidate builds introduces features like "the OrangeFS distributed file system, support for the USB 3.1 SuperSpeed Plus (SSP) protocol, offering transfer speeds of up to 10Gbps, improvements to the reliability of the Out Of Memory task killer, as well as support for Intel Memory protection keys," [according to Softpedia].
"Moreover, Linux kernel 4.6 ships with Kernel Connection Multiplexor, a new component designed for accelerating application layer protocols, 802.1AE MAC-level encryption (MACsec) support, online inode checker for the OCFS2 file system, support for the BATMAN V protocol, and support for the pNFS SCSI layout."
Just this morning, as I was reading an article on MS 10's latest update, which includes more adds, I was pondering; is it finally time to give linux another chance?
I give it a shot every couple years in hopes that I can finally ditch Windows, but invariably, I format the drive and go back to windows.
Now, to be honest, I did try Mint a couple weeks ago running on a live USB disk. It was reasonably snappy and looks "ok". However, there are definitely still big issues with High resolution displays, so that the scaling looks all wonky. Some elements look ok, but the icons on the launcher bar are so tiny I could hardly see them.
But... I could live with that.
What I cannot live with is the lack of games. Real games. New release titles.
I am hoping that can and will change with the release a Vulcan, though only time will tell.
As I get old though, the biggest issue for me is the lack of MS office. Yes I know there are open source alternatives, but these are not the same. I know people who are not big office users same it can be a one to one swap, but it cant.
If MS released a linux version of office, I would switch to Linux as my main OS and just boot to windows when gaming.
Just like clockwork... ... introduces features like "the OrangeFS distributed file system, ...
Just as I thought! I think we all has our suspicions, that Linux was an evil plot by the government, but here we see the proof: "Clockwork Orange"!!! Soon we will be forced to submit to the Ludovico Technique! Run for the hills.
One bit is very interesting to me:
A significant redesign to CPUFreq and P-State for allowing the kernel's scheduler to better communicate changes to the CPU frequency scaling drivers
Source: http://www.phoronix.com/scan.p...
It used to take some 30 ms for Intel CPUs to turbo-boost from a power-saving state (P-state). For CPUs in laptops, like the Core M series, this was noticeable when gaming. The latest-gen CPUs (Skylake) support very quick (1 ms) switching between P-states, and from what I gather, this kernel version now supports this. This means slight power savings and quick reaction from-and-to powersaving ("race to sleep").
Apparently it's very hard to get this right, because from what I read, the Microsoft surface tablets had a lot of trouble in this area.
8 of 13 people found this answer helpful. Did you?
Because allowing random userspace processes to mess with DMA capable hardware and PCI configuration space is a recipe for disaster.
A lot of hardware support is added in a new kernel. That's why sometimes, simply installing or upgrading to a new kernel gets rid of a number of problems, or improve performance. No wrong in this.
Slashdot, fix the reply notifications... You won't get away with it...
No it doesn't. You get a binary from Windows Update, or if you're lucky a setup.exe from the vendor.
GNU/Systemd, you mean.
OrangeFS is a fork/continuation of PVFS, a filesystem for high performance computing clusters. I understand it's rather popular in that world, as is Linux itself.
When, if ever, will this kernel make it to Ubuntu 16.04 LTS? Ubuntu 16.04 LTS uses Linux kernel 4.4.
Just like clockwork, the Linux 4.6 kernel was officially released today.
Clockwork was also released today? Damn! I missed that...
Garry Knight
Please, for the love of $deity, lets this be true. We've been putting up with broken video on, well, just about every Intel GPU since they stated their driver update for Gen9 (Skylake). And that includes older hardware that used to work before this effort was started. I can understand the occasional glitch in a new kernel, but "doesn't boot into X, at all, ever" isn't just a glitch - and it's been going on for 5 kernels so far. Currently in 4.5 I can't reliably attach a second monitor.
What amazes me is this isn't just Linux. The net was full of people complaining the video their brand new Windows laptop ranges from slow to utterly unusable. Naturally they said are going to get it fixed under warranty. Ha! It infests everything. The BIOS on my laptop can't initialise a second monitor either.
It is getting better. 4.2 didn't boot for me. 4.5 works acceptably on one screen. The i915's bugzilla reports my current two monitor problem is fixed. Hell, maybe I'll be able switch on full GPU power saving in 4.7! But is it really this hard?
Juicy details? Do you have some links/further reading for that?
I'd be genuinely interested: I collect those muddy things which, taken as a whole seem to be a "chaotic strategy" to keep real competition off (decommoditizing protocols et al).
Having slightly incompatible versions of very complex software "out there" and letting the (captive) market sort out the intersection of (unwritten) specs looks like a very smart way of keeping a compatible market from evolving
Plus: plausible deniability! ("your honor, we're just too stupid to come up with a clean design document and following through"). And the developers "in the trenches" even don't have to be aware of it!
Which loads a driver into the kernel.
Oh, you didn't know you don't have to compile *everything* in to the kernel? That's bad, mate.
Most of the driver-related stuff is in modules by the time it makes it to being installed on your system; but those drivers that are developed in cooperation with the rest of the kernel development process are referred to as 'in-kernel'. There are devices for which the drivers are handled out-of-kernel; but that's usually a world of pain.
Barely on version 4.6? He'll Microsoft was on version 2000 sixteen years ago!!! The lunix will never catch up at this pace.
Compile your own kernel, have only in there what you want. Be a luser (yes, this is the correct spelling) and have everything in there. As the kernel in in practice very modular, having this stuff as an option does not slow the rest down or cause problems. So far, this model has proven far superior to what MS does with their user-installed drivers.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Systemd is not required by the kernel and, for example, Debian still runs well without it, and I expect it will stay that way. The push-back if they tried to make it unusable without systemd would probably fracture the project and making it the default init-system was hopefully a bad enough experience for the systemd-mafia. It is really advisable to not use systemd on anything needing stability or security and I doubt it will ever mature to the degree needed to fix that. The design is just not clean enough and its main designers do not even begin to understand why KISS is critical in solid engineering.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
OSX is already up to version 10.11.5. These Linux people better start producing in order to catch up with the quality that is OSX and iTunes. They probably aren't even Agile. I'll mention it at the next stand up.
From uni users had guest lectures slides display different, crash the program, and have even had missing/non playing embedded elements(yes different person for each) ,but this is from ~3-5 years ago so who knows for the latest version. Fonts are still an issue (different ones on each system) and even if you take care to choose a cross platform font positioning will change due to different Kerning[1], this is an issue for most cross platform programs that rely on system font layout engines, but it is a design feature and so will not be corrected, if it was open source we would describe the bug as "closed wontfix".
[1]https://en.wikipedia.org/wiki/Kerning
It is really advisable to not use systemd on anything needing stability or security and I doubt it will ever mature to the degree needed to fix that.
Advisable by whom? And can I expect a well thought out and real reason for it other than "OMG systemd is big!"?
Meanwhile RHEL, CentOS, and such are still on 2.6 (eg from 2003, more specificly 2.6.32 from 2009)
The Authors clear lack of understanding regarding UIDs and TTYs (and the function of su et al).
Binary logging and interfaces with no explicit compatibility between versions.
Massive attack surface, scope creep and "single point of failure/pwnage."
Keeping EFI storage mounted read/write for the entire duration of the boot.
Now, Give us a well thought out explanation of why its BETTER than existing init.
No, they're referred to as "in-tree".
Kernel modules load into the kernel whether they are developed, tested, and distributed in-tree or not.
Its not time.
The OrangeFS page is so shitty they can't even explain what the filesystem is used for. Go ahead and see for yourself.
Only the State obtains its revenue by coercion. - Murray Rothbard
Is it as bad as it sounds, or something completely different?
Or, you could just do make localmodconfig; this only builds the modules you currently have loaded. You could make local yes config and simply do away with modules altogether, if you are sure that your HW setup isn't gonna change much. My experience is that most situations with USB are covered pretty well by the usb-storage modules and everything else can be builtin. Just to give more options to keep it lean-n-mean
C|N>K
Could somebody mod parent up?
Orange was a lemon for my workload, but it's fast if you can adapt your IO to use larger reads.
Orange also ate my data twice, making it useable mostly as a Job specific high GB/s fast scratch.
OK so this is just so Batman can remotely control the batcave from his Nexus 7?
YAY! I always wanted more DRM.
I have to admit that I haven't really spent a lot of time delving into the details ... but I recently bought a Dell XPS 13 9350 (Skylake core i7 CPU) with 16GB RAM and a 512GB SSD. Got a really good deal on it during a sale for Costco members and couldn't turn it down.
I'm wondering if there's a Linux distro that really works well with this yet, or do I need to give it more time?
I assume this new kernel was a piece of the puzzle for it, but I also heard there were issues with support for some of the features on the laptop. Is the built-in webcam working yet? Any audio issues? How is the 4K display w/touchscreen support?
The reason I moved away from Linux on any of my laptops was the common situation where "It works great except X, Y, and Z are broken." I don't want surprises like it not waking from sleep properly or the wi-fi not working or what have you. To me, those are deal-breakers vs. just using Windows on it instead and having everything function as intended.
I do non-module kernels with only the needed stuff for all production machines, because there have been module-loading related vulnerabilities in the past and this way I am not dependent on the user-space or an initrd for booting. I also do not use an initrd, and with systemd-free Debian that works still well and has for the last 15 years. Just do not do a kernel-package, but make it directly and boot it yourself. Reduces complexity considerably.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
You are free to use an inferior solution any time you want to. Just do not try to push me to do so. The technical aspects have been discussed to death and that you even ask means you are posting in bad faith.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Good list. However the systemd-fanatics cannot give you that explanation, they must attack what came before (and worked well), because they have no valid arguments why their way is better. That is why they universally use this tactics.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Actually the reason I ask is because the technical aspects have stacked in favour of systemd and much of that list is covered in a standard "idiots guide to clueless systemd complaints".
Also I'm not forcing you to use systemd. No one is forcing you to use anything. That's the lovely part about Linux. It's not like there's only 2 distributions out there. Go use whatever you want.
Also the way you automatically say inferior solution without understanding remotely my application or requirements means you are posting in bad faith.
Crap list which has been debunked. There are plenty of valid arguments for a change in init system. I mean it's not like there haven't been 10+ project who have done so, and it's not like the most popular Unix based distribution in the world changed its init system to be something very similar to what systemd is now with it's very first released version.... oh wait that's exactly what happened.
But hey I got exactly what I expected:
"Author bad and dumb mkay."
"OMG my logs and I can't read the manual to re-enable text logging" - cries
"OMG systemd is big!" - something which I specifically asked not to be included because the actual attack surface that is system critical is actually tiny. But the AC would know about this if he bothered looking into it.
"systemd mounts something that should be rw as rw and some vendor had a bug that once caused a problem on their machine, oh noes!"
Yep this is exactly why no one gives a shit about the anti-systemd community. I thought I was going to get something new or relevant. Alas.
It's fast and simple
http://linux-hacks.blogspot.in...
https://linuxconfig.org/buildi...
Casteism