Debian Warns Of Hyper Threading Issue With Intel Sky/Kaby Lake CPUs

[BaronHK]

Driving right by in my AMD Bristol Ridge APU laptop and desktop and looking at the carnage.

AMD has had nasty bugs too. Remember the original Phenom? They disabled TLB and made the processors slower, like, by double digit (10-20%) percentages.

So you had a choice. If you weren't running into the problem, you could choose not to update your BIOS, and if you were affected and wanted to do something to avoid triggering it, it was like a major processor downgrade.

You can read about the bugs that end up affecting AMD processors because they publish them in Erratum. They're really intended to document the problems for people like Linux kernel developers, so they fly under the radar usually, but there are a lot of them.

Like with Intel, most can be fixed, but only if you run newer firmware, which Debian won't include. So under Debian, you're basically running with the firmware that the BIOS loaded when the computer booted up, regardless of if it's broken and by how much.

Some of these bugs end up creating security vulnerabilities, so even if you run a fully patched Debian, without the firmware updates, you aren't getting the fixes applied to the firmware by the hardware vendor.

[starshipeleven]

That's kind of the point here, they already have a fix, whereas AMD don't even seem to have a clue why their CPUs don't work.

Well, if you keep serving the same reheated soup you're going to know it much better than those that try to bring a totally new design on the table.

[torsionbar28]

That's kind of the point here, they already have a fix, whereas AMD don't even seem to have a clue why their CPUs don't work.

^ And that folks, is how you spot an intel fanboi. Article about intel's processor bug, and all he can come up with is some baseless dig against AMD. Let me guess, you paid waaay too much for an i7, right before intel cut prices as their response to Ryzen?

FYI both intel and AMD publish erratum documenting flaws discovered, and microcode/BIOS updates for fixing said flaws.

[Jumbotron]

AMD has had nasty bugs too. Remember the original Phenom? They disabled TLB and made the processors slower, like, by double digit (10-20%) percentages.

So you had a choice. If you weren't running into the problem, you could choose not to update your BIOS, and if you were affected and wanted to do something to avoid triggering it, it was like a major processor downgrade.

You can read about the bugs that end up affecting AMD processors because they publish them in Erratum. They're really intended to document the problems for people like Linux kernel developers, so they fly under the radar usually, but there are a lot of them.

Like with Intel, most can be fixed, but only if you run newer firmware, which Debian won't include. So under Debian, you're basically running with the firmware that the BIOS loaded when the computer booted up, regardless of if it's broken and by how much.

Some of these bugs end up creating security vulnerabilities, so even if you run a fully patched Debian, without the firmware updates, you aren't getting the fixes applied to the firmware by the hardware vendor.

Uhh....you did notice I said Bristol Ridge....not Phenom.

Also if Debian is that shitty.....stop using Debian. Open source.....means choices. Make a new one and move on.

[VikingGe]

^ And that folks, is how you spot an intel fanboi. Article about intel's processor bug, and all he can come up with is some baseless dig against AMD. Let me guess, you paid waaay too much for an i7, right before intel cut prices as their response to Ryzen?

Interesting accusation there, but no, I'm on a Phenom II X6 that I'm probably not going to upgrade before next year.

You might have missed it, but this little problem is still open, which is what I was refering to. Affects quite a few people, exact cause apparently still unknown, no fix in sight. And that is exactly the problem, I'm certainly not going to buy a CPU that is known to be unreliable in an everyday task.

Before you ask, yes, I did say basically the same things about Skylake when the random freezes with Prime95 were discovered. Took Intel a while to fix that as well.

[pdffs]

Wow, thanks for the heads up! The updated intel-microcode package they recommend doesn't exist in the Mint 18/Ubuntu repos yet, but I can confirm that with the "iucode-tool" dependency installed, the .deb package from debian installed just fine on Mint 18. Haven't rebooted yet though, running a full backup before I do.

The intel-microcode package is available in the restricted repository, which is not enabled by default (check /etc/apt/sources.list, or your GUI package manage to enable)

[waxhead]

It's funny how Debian goes to all this trouble to be "Free", but then they package things like Widevine, Flash, RAR, etc. and just say that it's not officially part of Debian.

No it isn't. Most of Debian is free and it works for most people out there. Yes, there are some trouble spots with some hardware, but if you are having a problem with Debian being as free as possible you don't mind spending a few bucks on hardware that is known to work do you?! You should appreciate that someone is trying to make stuff free for us all instead of talking crap about it. Thanks to large distros like Debian trying hard to get things right other distros benefit from that which again makes it possible for you to run a free operating system where things just work out of the box. Your argument is a bit like censoring half of Wikipedia unless you pay for it.

[BaronHK]

No it isn't. Most of Debian is free and it works for most people out there. Yes, there are some trouble spots with some hardware, but if you are having a problem with Debian being as free as possible you don't mind spending a few bucks on hardware that is known to work do you?! You should appreciate that someone is trying to make stuff free for us all instead of talking crap about it. Thanks to large distros like Debian trying hard to get things right other distros benefit from that which again makes it possible for you to run a free operating system where things just work out of the box. Your argument is a bit like censoring half of Wikipedia unless you pay for it.

They ship without firmware and then stand there like the retarded stepchild when computers malfunction because of it. You think you're not running "non-free microcode" because you don't have the Debian package installed? WRONG! You're just running an old version with bugs as loaded by your BIOS! Debian's policies mean that unless the user configures their computer the way Debian should have done it, then this FIXED bug could be provoked and result in a disaster complete with data loss.

Review of Debian 9: Broken Broken Broken Stupid Stupid Stupid. Full Stop. There doesn't exist a universe where "We're going to leave you with the equally non-free broken BIOS-provided firmware instead of shipping the patch." makes sense.

Which x86 processor that anyone other than Richard Stallman would want to use doesn't have non-free firmware?

I know that Debian supports a lot of architectures that aren't very important in the grand scheme of things (I mean, up until recently, if you wanted to dust off your PowerPC Mac from when dinosaurs roamed the Earth and have Debian stop to fix issues with it as if it was as important as x86, you found your distro), but probably upwards of 90% use it on x86, and most distributions don't even support more than x86 and maybe a few arm devices. Fedora has even de-prioritized 32-bit x86 to the point where problems that only affect that version don't get to block a release.

[MaxToTheMax]

The intel-microcode package is available in the restricted repository, which is not enabled by default (check /etc/apt/sources.list, or your GUI package manage to enable)

Yeah, I could find an intel-microcode package, but it was clearly too old (released in 2015.)

[wizard69]

intel regularly releases new microcode for their cpu's. over the years intel has had a lot more misbehaving silicon than amd.

Haswell & Broadwell:
http://www.anandtech.com/show/8376/i...eep-broadwelly

Core2:
https://www.theinquirer.net/inquirer...ore-cpu-errata

Another Skylake bug a year ago:
http://www.bit-tech.net/news/hardware/2016/01/11/intel-skylake-crash-erratum/1/

This is so true yet every time something like this happens Intel gets a pass. I'm not sure what the issue is but over the years I do believe AMD has had fewer problems with its processors.