![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
Intel Discontinues Clear Linux
- Thread starter TurboGLH
- Start date
- Joined
- Nov 7, 2008
- Messages
- 5,962
Aww, that's a shame....
I remember when I first tried it, I was getting like sub 2 seconds boots looking at systemd-analyze blame. Was interesting to restart and you're back at the login in the time it took to blink.
Package manager was a pile of shit.
Never really saw it go anywhere in the container space.
I don't really know what it did well besides being fast.
Package manager was a pile of shit.
Never really saw it go anywhere in the container space.
I don't really know what it did well besides being fast.
A lot of other distros have borrowed a lot of clear tweaks over the years. It is sad to see it go... but Intel can't afford the man power on it I guess. I just hope they are refocusing most of their decent Linux folks and not just axing them. IMO that would be a massive mistake. Intels own tweaks to the kernel and other core systems was about all that has kept their servers semi competitive. If the handful of Intel people responsible for all those commits are gone with Clear its not good for them long term.
Intel should never have caved to MS back in the day and gave up on their own consumer focused software projects. World would be different if Intel had told Bill to fuck off back in the 90s.
DukenukemX
[H]F Junkie
- Joined
- Jan 30, 2005
- Messages
- 11,195
Clear Linux did it's job, and that's to prove you can get faster results with optimizations. Nowadays you can achieve the same or better performance with other distros. Also, nobody realistically used Clear Linux.
https://www.phoronix.com/review/framework-12-linux-os/7
https://www.phoronix.com/review/framework-12-linux-os/7
https://cachyos.org/blog/2409-september-release/
gcc: Cherry-picked patches and flags from Clear Linux
glibc: Added “evex” patches as well as cherry-picks from Clear Linux
Clearly not all the performance upstream is Intels doing... but a lot of it has been. They just upstream all their work. 4-5 years ago Clear was 20-30% faster then everyone else. Intel has up streamed a ton of performance tweaks that are just part of the kernel these days.
Intel has been responsible for 7-17% of Linux code every single year for over a decade now.
They don't have to have clear linux as a distro to continue that. I am just saying I hope the 20-30 people that have been responsible for 90% of that Intel contributed code are not on the way out the door with clear. For all the shit I give Intel, they have been good about spending some money on Linux development. Having guys of the caliber of the Intel employees that were on clear is NOT cheap. If they are out the door they will all have jobs in Linux land somewhere within a week. I just doubt it will be here is your paycheck make Linux rock kinda jobs.... most companies pay folks like that to do more specific benifit them sort of things. How many companies pay engieeners to do something like Clear, and have all their work make AMD servers better?
Last edited:
It wasn't supposed to interest you. They didn't really want people using it as a daily driver, or a server distro. It was a platform for them to test on. They were not looking to compete with Microsof, or red hate or ubunut.
Most big companies have their own internal distro spin. Microsoft has their own internal distro that you can't download and use. AMD has an internal distro as well....
Intel was just odd in that they allowed anyone to download and use their internal testing Linux distro. If your going to have a team of people working on Linux stuff you sort of all need to be on the same base.
https://www.phoronix.com/news/Clear-Linux-Popular-Recap
I gotta say it feels like piling on Intel to joke about it at this point. But I can picture the new CEO going through a list of things to axe... seeing this and saying yep ok that's gone. lol
I gotta say it feels like piling on Intel to joke about it at this point. But I can picture the new CEO going through a list of things to axe... seeing this and saying yep ok that's gone. lol
Can you cite this because I cannot find anything supporting it.
I find it hard to believe they would opt out by default. Maybe, but I don't really see why they wouldn't just use auto.
Kind of weird because some of those exploits can lead to nasty things like a VM Escape at worst.
Clear is definitely not in the lead purely from that, at least.
Edit: Or did phoronix fuck with it? Which is just bad benchmarking.
Last edited:
By default Clear has(I guess had) all the current upstreamed CPU security mitigations. I'm not really sure why pioruns believes they were disabled by default. They were not.
I'm not really sure where the idea that clear wasn't a secure OS comes from. As far as I know none of Intels compiler tweaks (many of which have been adopted by distros like Cachy) were ever bypassing security fixes. Intel wasn't selling it as a production ready server OS or anything. It was a test bed for sure... but the version you could download and install yourself was no less secure then any other Linux distro. Obviously it wasn't a hardened server like SUSEs Sles or something... but if you did use it it wasn't insecure. Annoying to use Intel didn't care about that sure... but not insecure.
Linux has always been able to disbale kernel level security fixes... it was kernel 5.2 that added the mitigations=off boot variable to disable all of them... and that was added by Josh Poimboeuf from Red Hat.
As far as I know Intel has never had a hand in any of the earlier kernel level by pass flags like spectre_v2_user=off either. I may be wrong on this one but I believe that earlier flag was added by Andrea Arcangeli also from Red hate and I believe at one point SUSE.
https://www.phoronix.com/news/Linux-Spectre-SECCOMP-Default
Intel didn't have to add any of the mitigation off stuff themselves to Linux... all the server running, and building companies wanted to have a way to turn them off.
Anyway to my knowledge everything Clear did to get more performance was above board. If it wasn't why would it actually have benefited AMD more then Intel? lol I mean if there was shady disabling of Intel specific security fixes going on... why was clear the fastest Linux platform for chips like the AMD 9950x?
Like Cachy... Intel was custom compiling all the core files for extended extensions. Things like AVX-512. Which ironically is probably 3/4 of the reason why Zen5 performs so well overall on Clear.
Last edited:
I always thought Clear Linux run with mitigations off, even stupid ChatGPT lied and confirmed this when I asked, just before writing my first post. Looks like opposite is true. My apologies, I deleted my two posts in this thread.
No need to delete. it is crazy how often Google searches are returning chatgpt powered bs. It's one thing to get crap search results but the terrible ai summery stuff is close to criminal.
Yes, ChatGPT is an abomination. Lying through its teeth. Google AI summary thing isn't any better.
WOW ya holy S that is a bold face. Lie. I wonder if it does that sort of things with commercial distros. Clear isn't an actual commercial product. I could imagine if it was saying that sort of thing about something like RHEL a company like IBM could sue them for that sort of thing.
EDIT: I googled "Does SLES linux install with mitigations off"... and it said. "Yes, SUSE Linux Enterprise Server (SLES) allows users to install with mitigations for CPU vulnerabilities either disabled or enabled during installation. "
At first I was like woo wtf. Then realized ok ok no SUSE does give you a checkbox option at install to disable mitigations. haha
I couldn't immediately find this clearly (haha) documented, so yeah, I imagine it just pulled that out of its ass based on other context.
Migrations = slow.
Clear Linux = fast.
So it made the logical leap that they're probably disabled.
- Joined
- Apr 5, 2007
- Messages
- 4,574
I could not care less if a distro disables Mitigations or not by default (I don't know any mainstream distro that turns them off by default). You should always make sure they're enabled based on what the device is doing. Assuming they are in place will always get you into trouble. A server with a job that may involve the cloud or hosting a site of some kind? Keep them enabled. A Linux admin that doesn't understand that sort of thing shouldn't be an admin.
My personal laptop running Arch? Yeah, I have mitigations off because I'm not in a position to be compromised with them off. So I'll take my performance boost thank you. All my servers running my many self-hosted tools? Mitigations are on even though they're not open to the Internet.
Still sad to see Clear Linux going bye bye. The work Intel did for performance was great for Linux as a whole.
My personal laptop running Arch? Yeah, I have mitigations off because I'm not in a position to be compromised with them off. So I'll take my performance boost thank you. All my servers running my many self-hosted tools? Mitigations are on even though they're not open to the Internet.
Still sad to see Clear Linux going bye bye. The work Intel did for performance was great for Linux as a whole.
It would have worked well... I mean all the packages are out of date already now.
If you need a realtime kernel now I would look at a distro like Cachy/arch and run a realtime kernel.
If you aren't talking about a server... but using realtime on a desktop/workstation for like audio production or something. And interesting newer option is to use the sched-ext user space CPU schedulers. scx_flash is very low latency, and has a tickless mode. Advantage to using a ext scheduler is of course being able to turn it off when you don't need the improved latency.
Isn't the special sauce getting merged into every mainstream distro what normally happens any time someone spins up a new distro to do one new thing that no one else has yet?
- Joined
- Jul 1, 2025
- Messages
- 91
It is normal if chatgpt searches the web to get updated information. But it is bad when google relies on chatgpt to get result, it will be fake especially when it is about contemporary topics which are not updated in chatgpt database yet.