/tech/ - Technology★

>>856426

Why'd ya sage?

fuck off back to 4chan

>>856430

>implying that this isn't news to all people

Linus basically shut down Intel's proposed Linux patch which would slow the shit out of everybody on every architecture.

Keep on cryin'.

it's nice to see the goy finally let some steam out after sucking dick during the embargo, hehe he sure showed them

What happens when linux is gone? Will CIA niggers instantly ruin linux?

>>856436

linus*

Interesting, but while the ranting is nice, we're not left with ANY alternatives.

>>856438

The fact that Linus Torvalds is barking at Intel like that is something in itself.

linux about to start blacklisting intel's microcode updates

> Oh yes, we want a microcode blacklist. Ideally we refuse to load the

> affected microcode in the first place and if its already loaded then at

> least avoid to use the borked features.

>

> PR texts promising that Intel is committed to transparency in this matter

> are not sufficient. Intel, please provide the facts, i.e. a proper list of

> micro codes and affected SKUs, ASAP.

http://lkml.iu.edu/hypermail/linux/kernel/1801.2/04521.html

>>856447

http://lkml.iu.edu/hypermail/linux/kernel/1801.2/04979.html

On Sun, Jan 21, 2018 at 03:56:55PM +0100, Borislav Petkov wrote:

> Also, blacklisting microcode for early loading will become an ugly dance

> so I'd like to avoid it if possible.

>

> Thus, it would be much much easier if dracut/initrd creation thing

> already filters those blacklisted blobs by looking at the revision in

> the header. Which is much easier.

>That wouldn't be enough; AFAIU there's people >with this stuff already

>flashed in their BIOS. So the kernel needs to deal with it one way or

another.

reminder never to do bios updates, they are infected with botnet.

>>856447

>linux

I'd just like to interject for a moment. What you're referring to as Linux, is in fact, GNU/systemd

Now you can have your botnet init load botnet microcode blobs on to your botnet CPU at each boot.

>>856438

The solution is to fix Intel's shit, which slows down Intel by 30% or more. Intel's proposals would slow everybody down by that much.

Linus just BTFO'd Intel because they're playing games like this.

>>856449

>>856447

ahahhaha holy fuck this is priceless

it's spooky to think what will happen to gnu and linux once their monolithic figure heads eventually bites the dust

>>856438

The alternative is to use AMD instead.

>>856496

99% of the work done with Linux in the last 20 years has basically been for corporate customers. As a Unix clone it's been good enough for that entire time.

The same can be said of GCC, the most important GNU project.

>>856497

Meh, you should at least go ARM A7 if you're a poorfag, or POWER if you got cash.

Well anyway I'm sure all the gaymers will keep finding excuse to stay on x86, but whatever. Not my problem.

>>856498

OpenBSD despite being cucked hard on the license front is the least cucked when it comes down to actual code.

>>856498

The work is mostly done by corporate customers but it's still gatekept by Linus. He has prevented a lot of bullshit.

>>856514

BSD licensing has a nice side property to it, you don't get corporations trying to push pure shit back into OpenBSD, they just silently fork and do retarded things on their own time.

>>856531

> they just silently fork and do retarded things on their own time.

That's very subtly powerful argument against the GPL virus.

>>856548

who funds the jews?

its got to be one person becuase that hair style is rediculous and nobody in their right mind would use it unless they fully understood the game

> And that's actually ignoring the much _worse_ issue, namely that the

> whole hardware interface is literally mis-designed by morons.

> It's mis-designed for two major reasons:

> - the "the interface implies Intel will never fix it" reason.

epic

>>856547

Daily reminder that the GPL does not require contribution back into the upstream projects.

>>856436

With all the luck, by that time HURD will be usable. I think there were a bunch of others neat kernels, GNU OS can swap them around pretty easily.

>>856623

Daily reminder that the GPL is not about getting contribution back to the upstream project

>>856636

Daily reminder that according to the GPLv2

The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable.

it's not the source code if it's intentionally obfuscated and comments censored under the guise of an "embargo".

>>856637

https://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html

3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,

how can linux distribute obfuscated code instead of the source code and still keep the GPLv2?

>>856638

It says "machine-readable" so I'm afraid that includes obfuscated shit as well. Frankly it should have been "human-readable" but I guess that would preclude the use of binary data.

Can't you license wankers make your own containment thread?

>>856635

Realistically, we can already replace Linux with any of the BSD kernels. We'd finally have proper ZFS support too. GNU/k*BSD, as I like to call it.

>>856648

>cuck license

>>856650

Can be forked into GPL, if need be. And most big companies treat Linux like it was licensed under the LGPL anyway.

>>856514

OpenBSD doesn't give a fuck, AFAICT they reject a shitload of patches in favor of correctness anyway so I don't think the license hurts them. If it does we'll never know.

>>856531

>The work is mostly done by corporate customers but it's still gatekept by Linus. He has prevented a lot of bullshit.

This. Linus operates similarly to the chans in that he seems to actively discourage participation from morons by ragging on them.

>>856638

Use an obscure format? Hey, here's your source code, on unobtanium WORM drive platters / DATA-MD / 7" floppy / DEC disk packs / punchcards / etc. For only the cost of media plus shipping!

>>856676

>on a medium customarily used for software interchange

If it's an obscure format, then it cannot be a medium customarily used for software interchange.

Jews will find a hole in even the GPL if they can, because they're jews. This does not mean the GPL is useless. The harder it is to find the loophole, the less "people" can get away with violations. That said, if nobody actually takes the violators to court, then it does not matter what license you choose. It's just as important to enforce the license, no matter how strict or how cucked.

>>856432

This isn't news to anyone who gets their news from more than just /roachpol/. Fuck off back to /g/

>>856509

POWER + x86 on an expansion daughter board for backwards compatibility is the way forward.

>>856684

Language should be changed to contemporarily.

>>856684

Language should be changed to contemporarily.

>>856873

Would PCI-E be enough or do we need a new socket with a different transport layer?

Are there anyone actually dumb enough to buy Intel, after this month? We can all hope that Intel's propaganda campaign at confusing the working Meltdown vulnerability and the more theoretical Spectre vulnerability fails, but knowing how stupid most people are, they are going to keep falling for Intel. Let's do all we can to destroy Intel's reputation once and for all.

>>856895

Certainly people are dumb enough to buy Intel. Tee Vee isn't telling them not to.

Never underestimate the stupidity of the general public anon.

>>856896

If Intel can do propaganda against the thruth, why can't we play equally dirty for the truth?The best thing we can do now is to encourage people to buy AMD. I know they have unfortunately included the PSP on their newer processors, but they are still better than Intel, and until every poorfag can afford a Talos Workstation, the best choice. A normalfag requires a simple alternative, something that he can get here and now. We should also inform them how Intel is the Microsoft of the CPU world with their fraudulent (((salesmanshi))), and cheating compilers.

>>856919

Frankly I want people's computers to fail en-masse. It would please me. We tried to warn them and they looked away. Now it's time to pay the price.

I do try to warn people but their eyes glaze over. Why waste the effort?

I do commend you if you have the youthful energy to keep pouring your heart into trying to wake up the sleeping masses but I just don't give a shit any more. It's not worth it to me. Cast not your pearls before swine.

>>856873

>POWER + x86 on an expansion daughter board for backwards compatibility is the way forward.

We are reliving the 1980s and I am OK with that.

Duel CPU's where common back then. Apple2's where 6502 but you could get a Z80 card to run both software types. The C128 had a 65xx and Z80. Tandy had machines that did the same thing.

And there where the "DOS Compatible" Macs

>http://www.pcmuseum.ca/details.asp?id=108

> Apple also released a PC-compatible model of the 6100 called the Power Macintosh 6100 DOS Compatible. This version came with a card that featured an Intel 80486DX2 processor and a single SIMM RAM slot that used a different type of RAM to that in the Power Mac itself, and also sported standard PC VGA and joystick ports. One could easily run the Macintosh interface and DOS/Windows 3.1 side by side, even on different monitors if one so desired.

They also had "PC personality cards" for the later PPC macs. Was a Pentium and ram on a PCI card.

>>856635

>HURD

>usable

>ever

>>856868

Torvalds has said time and time again long before the Linux Foundation that he doesn't care about politics. He has explicitly dedicated his life to being apolitical in his work as a software engineer.

>>856884

New bus to avoid the bus niggers, or something like HP-IB but higher speed.

>>856684

Language should be changed to contemporarily.

So if you dramaqueens actually followed the thread you would realize Linus misunderstood the patchset proposed by Woodhouse.

>> The patches do things like add the garbage MSR writes to the kernel

>> entry/exit points. That's insane. That says "we're trying to protect

>> the kernel". We already have retpoline there, with less overhead.

>

> You're looking at IBRS usage, not IBPB. They are different things.

To mitiate Spectre variant 2 both Intel and AMD added three new features on microcode. IBPB, STIBP, IBRS.

IBPB : discard all branch targets learned earlier.

STIBP : prevent hyperthread sibling from following branch preditctions learned from another sibling

IBRS : prevents branch target learned in a less-previleged execution mode (before it was most recently set) from takin effect. Needs to be set on each kernel entry FOR REASONS.

There should be no reason to rely on these features if everybody uses bleeding edge gcc version with Google's repoline support but unfortunately Skylake will be still vulnerable even with retpoline under deep call stack FOR REASONS. That is why Woodhouse wanted to use IBRS on Skylake processors on his last patch set but Linus was too tired to understand what the fsck was he trying to achieve.

Seems like there are other solutions to save skylake cucks by tracing the stack depths from the kernel side and whats not. Might be good chance to follow their conversation and patchsets if anyone's interested in learning how linux handles process context switches on hyperthreaded processors.

>>857107

Thanks for that friend. /tech/ is not exempt from confirmation bias.

>>857107

>all this shit

how about intel just stop producing processors with botnet and recall all processors with botnet and provide replacement processors with no botnet.

all intel kikes can burn in hell until this happens.

>>856896

I built a new PC with a 6700k a year ago. I really regret not waiting for AMD's new processors.

>>857111

>/tech/ is not exempt from confirmation bias.

YOU DON'T SAY

>>857107

>but Linus was too tired to understand what the fsck was he trying to achieve

Unironically Linux pls

>>856868

>2nd pic

Why do I get the feeling you have no idea what that actually means?