Linux 4.6 Brings NVIDIA GTX 900 Support, OrangeFS, Better Power Management (phoronix.com) 129
An anonymous reader writes: The Linux 4.6-rc1 kernel has been released. New to the Linux 4.6 kernel are a significant number of new features including NVIDIA GeForce GTX 900 open-source 3D support when using the closed-source firmware files, Dell XPS 13 Skylake laptop support, a fix for laptops that were limiting their own performance due to incorrectly thinking they were overheating, AHCI runtime power management support, Intel graphics power management features enabled by default, a new file-system (OrangeFS), and a range of other improvements.
OrangeFS (Score:5, Funny)
Developers provided performance charts against AppleFS in the release notes, but they found it wasn't comparing the same thing.
Re: (Score:3)
I looked over the homepage for OrangeFS and still can't figure out the point.
Re:OrangeFS (Score:5, Informative)
From a glance at their website, I think the keyword is MPI-IO. Picture a weather simulation consisting of 10,000 processes that work together in parallel, where all of them perform semirandom input/output using the same files. That's gonna be difficult to do right with conventional filesystems...
Re: (Score:1)
Your post was 1000 times more informative than their site. Thank you.
Re: (Score:2, Funny)
That seems like a crazy waste of energy. If you want to predict the weather approximately as accurately as the local news, just roll a d20 and look the results up in your DM manual.
Weather prediction (Score:1)
That seems like a crazy waste of energy. If you want to predict the weather approximately as accurately as the local news, just roll a d20 and look the results up in your DM manual.
But this only works correctly, if you don't use the cheap molded/ground dice, but get some machined ones.
Hollow machined dice, filled with the blood of a weatherman.
Re: (Score:3)
When people here filesystem they think of a disk file system.
Maybe calling it Orange Clustering System might be a more descriptive name?
Re: (Score:1)
"Picture a weather simulation consisting of 10,000 processes that work together in parallel, where all of them perform semirandom input/output using the same files. That's gonna be difficult to do right with conventional filesystems...'
But absolutely trivial to do on a RAMDisk. This seems like a solution in search of a problem, however said problem was solved long ago with far superior capability.
Re: (Score:2)
Re: (Score:1)
Do explain how to share that RAMdisk between tens of thousands of processes running on multiple different systems, possibly handling petabytes of data and needing redundancy, too? Oh, right, RAMdisks can't handle such needs.
If the same file is accessed by multiple nodes, this implies no data moves. That is, each part of the file has a specific meaning. When you're building a cluster for such a purpose, there likely is no need for multiple files. Or, for performance reason, noone sane would put multiple files on the same disk. Speed will be needed, and RAM can easily be build as big as today's SSDs, even server ones. What's needed is simple direct access to the same block device, with no file system getting in the way at all.
Re: (Score:2)
Re: (Score:2)
When you make a RAMDisk for multiple machine access, you give it no FS and just give it raw data with pointers to that data when requested.
It's exactly how my 2D Second Life clone works. Very efficient, very fast, all world changes are practically instantaneous. One raw binary resource file, multiple nodes access it all from RAM, every access forces a refresh across every accessing node. You only need a proper backplane for node interconnectivity.
Re: (Score:1)
"And no i am not saying that ram i slower but when you add a disk emulation layer it is."
You are still wrong.
If you did indeed observe a RAM disk to be slower than an SSD, then you fucked something up.
The point is to have yet another thing (Score:2, Insightful)
Seriously, there seems to be an obsession in parts of the Linux community to make yet another thing that already does what a bunch of existing software does. I'm not sure why but you see a lot of it. You'll get a distro that'll have 8 different media players, none of them worth a damn, rather than just having one good one but hey, you have options!
Re: (Score:1)
Looking at his code in systemd.... Shitting is pretty accurate.
It blows my mind that Linus allows that no talent hack to put anything near the kernel.
Re: (Score:2)
He also said "Linus allows", Linus only has control over what goes *in* the kernel...
Re: The point is to have yet another thing (Score:1)
So you think it's ok for Ubuntu to decide what you should be allowed or not allowed to install?
Re: (Score:2)
Re:OrangeFS (Score:4, Informative)
Re: OrangeFS (Score:1)
You know that orangeFS is a special clusterd fs and that not one of the filesystem that you mention would work as a sutible replacment.
You are comparing a hammer to a saw please consider the rigth tool for the jobb
Re: (Score:2)
Re: (Score:2)
The truth is supporting projects like this are more trouble than they are worth. No matter what they release someone is going to bitch and moan that something isn't correct. AMD/Nvidia can cater to their market which is 90% gamers and 9.9% professional graphics. The other 0.1% are foaming at the mouth zealots who demand source files but would never actually read the million or so lines they are composed of. I can't blame them for putting in minimal effort. How many extra card sales will open source drivers
Re:when is it going to be different? (Score:4, Informative)
I have a GTX 660. I have it because the folks who built my system for me included it as part of a package that had everything else I wanted. I was a bit leery of it at first, but went with it after finding that it was listed as supported by Nouveau, and that Nvidia provided its own Linux drivers as well. Nouveau sucked: it did not support a number of features that it claimed to, and was flaky as hell. Got tired of of my desktop vanishing without a trace, so I decided to try the Nvidia drivers, which worked a treat. Sometime later, after a number of successful driver updates, I hit one that didn't go so well. Filed a bug with Nvidia, and found their tech support folks to be extremely polite, knowledgeable, and helpful--and they actually took the time to explain a few things to me.
I prefer to use FOSS software. But I also prefer to use a computer that works.
Re: (Score:2)
Only if you believe that 98% FOSS is no less a failure as 0% FOSS.
Since I don't necessarily see everything in terms of black and white, I'll settle for the 98%, tick the Success column, and proceed to get some work done. House payments don't grow on trees, and my mom doesn't have a basement.
Re: (Score:2)
"Than", not "as". Sorry.
Re: (Score:1)
I prefer to use FOSS software. But I also prefer to use a computer that works.
Wow, why don't you just march right up to Stallman, pull on his beard, and tell him to go fuck himself. He fought long and hard for your freedom! Freedom from the evil slavery of proprietary software like Super Mario Borthers. You make me sick! God damn you!
Re: (Score:2)
I prefer to use FOSS software. But I also prefer to use a computer that works.
Wow, why don't you just march right up to Stallman, pull on his beard, and tell him to go fuck himself. He fought long and hard for your freedom! Freedom from the evil slavery of proprietary software like Super Mario Borthers. You make me sick! God damn you!
Hey now.. one cannot simply walk in and proceed to pull on the beard of Stallman. You insensitive clod. What gives you the right? That thing is his, you know?
Re: (Score:2)
Nope. His beard is open source. It belongs to all of us.
Re: (Score:2)
Re:when is it going to be different? (Score:5, Insightful)
This is what they've been saying about both AMD and NVIDIA since there was a linux and an AMD and an NVIDIA. They all say "open-source xxx with binary". So what, they are drawing the line somewhere else?
Sounds to me like an open source driver in the OS and an opaque firmware blob to be loaded into the peripheral and run entirely there.
Not ideal. But how (besides the complexity and ease of installing malware) is it different from doing a complex silicon design, with an open driver, and not giving the RTL description of the logic? Or doing an FPGA design, providing an open driver, but not giving the source to the FPGA load, only the opaque binary object that describes the logic to be emulated?
Re:when is it going to be different? (Score:4, Interesting)
Another Reason (Score:4, Insightful)
Re:Another Reason (Score:5, Interesting)
So there is a high probability that nVidia are taking the Microsoft coin and in return the deal prohibits them from providing full open source to the FOSS community...
nVidia has hinted around repeatedly that getting into bed with Microsoft and producing NV2A is specifically where they became massively encumbered. A lot of people who claim to be interested in this stuff don't seem to know that Microsoft was dipping their toe into GPUs back in the nineties with Project Talisman, mostly being done by Cirrus Logic with some input from Silicon Engineering, Inc. It had features not then in use by other graphics solutions, including a skewing technique that permitted you to interpolate some parts of some frames instead of actually fully rendering them. Who knows what patents were cross-licensed between nVidia and Microsoft?
Re: (Score:2)
While there "could" be the ex-SGI guys at Nvidia that got burnt in court before were pretty vocal about the one I mentioned.
Re: (Score:2)
The ideal is convincing Nvidia that software patents will not be an issue if they open up the code. We may have to wait for the ex-SGI guys in that place to retire because they were burnt before. The absolute ideal way for that to happen is if those stupid software patents that are normally just a description of a problem instead of a solution to be completely discarded.
There was an article on ARS T, that Intel is considering to replace NVIDIA for 2017 with AMD. It makes sense for me. Intel needs a less powerful second source and they also need to insure AMD is kept as 2nd source. They also like the AMD graphics hardware.
As we say, "We shall see!"
x86 Assembly rewritten in C (Score:1)
Is that a good thing?
Re: (Score:1)
In general: Maybe. It's customary to have a C version and have assembly versions "override" the C as and when needed. If all you had was the assembly version then a C version would likely be nice to have around. If it's a piece of code that only makes sense on one architecture anyway and the replacement runs half the speed and double the code size, that effort might well have been better spent elsewhere, even if the code wasn't performance critical.
In this case... well, it appears to be a maintenance issue,
Re: (Score:3, Insightful)
Debian bug #1558331 proves that. They claim MongoDB and Chrome are tools of the repukianz so they don't allow us to install them.
Wow, this shows the hatred of Debian contributors:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1558331?comments=all
They hate us and don't want to allow us to use the software we need.
moderators? (Score:1)
Why was this voted down? It included a link to the discussion of exactly why we aren't allowed to install why we want to install.
Re: (Score:2)
Re: (Score:2)
How can they stop you from using it?
Re: (Score:2)
I don't quite get how inventing time travel is off topic for a technology site. Well... actually I was hoping for +5 off topic, but the AC's comment has shown me the error of my way. I apologize. Sex is inappropriate,.. got it.
I swear, this time travel business is nothing but trouble.