Linux Stable Updates Are Dropping The Performance-Pounding STIBP

[dungeon]

It is present in 4.9 tree too as of 4.9.137, should be removed from there too

[lucrus]

Michael I'm no native english, but I feel like this sentence should be edited:

and thus the performance regressions restored

Maybe that should read something like:

and thus the performance regressions removed

?

[geearf]

Michael I'm no native english, but I feel like this sentence should be edited:



Maybe that should read something like:

?

Maybe just "performance restored", but yeah restoring a regression doesn't sound like a good idea.

[tomas]

Heck, Windows 10 runs 5 different distros with the click of a button.

WSL is Linux in the same way that Wine is Windows. The "distros" you refer to do not include the most fundamental part of a "Linux distro", namely the Linux kernel itself. WSL is simply an attempt of damage control from Microsoft to mitigate the flow of developers choosing Linux as their development environment. It has nothing to do with the famous "love" from Microsoft. That "love" is solely reserved for Linux on servers since it drives profits for their Azure business.

[cl333r]

It's statements like this because of which the Linux community will never be taken seriously. You guys need to grow out of this fantasy of Gates eating Linux developers' children in the black forest. It's immature, unjustified, and just plain stupid.

For starters, half the Azure VMs run Linux. .NET and PowerShell run on Linux. SQL Server runs on Linux. VS Code, the most popular development environment according to the latest Stack Overflow survey, runs on Linux. Heck, Windows 10 runs 5 different distros with the click of a button. Microsoft is a platinum (highest level) member of the Linux Foundation. Will you ever realize it's no longer 1995 and Microsoft no longer consists of just Windows and Office?

Will you?

Why are you crying like a little bitch? Is this your first year with internet?

[Weasel]

The more I contribute to various open source projects the more I realize the only sane project is the Linux kernel because it has the sanest maintainers. Decisions like this are perfect proof of that.

[dkasak]

What do you mean? Anyone can read or participate in the public mailing list discussions..

Not all discussions happen there. There would have been quite a bit of discussion with Intel techs and PMs that we'd never know about. Intel obviously came to the kernel devs with a relatively coherent position, after this internal process.

Are you aware that the actual mitigation is to disable hyper threading, and that the STIBP mitigation was not a complete solution? So the very security-conscious users who are using those affected processors would not even consider STIBP.

Yeah. Except that cloud providers like Amazon don't disable HT. Yeah you can disable it after boot, and as others have said, there is "always" a trade-off between security and performance. The thing is ... most people don't consider themselves *that* security-conscious to constantly track these issues and make drastic changes like disabling HT ... until the point at which they're affected in real-life. Personally I think the default configuration should always be the most secure. If people want to tweak this to remove security and gain performance, that should be an actionable option. But not the other way around.

[Weasel]

Personally I think the default configuration should always be the most secure. If people want to tweak this to remove security and gain performance, that should be an actionable option. But not the other way around.

Buy a CPU without HT then. It's the "default".

Most of the people you speak of (casuals) care more about performance than security anyway.

And if Amazon doesn't disable HT then it only shows they also care about performance more. You don't decide what's important for others. If you have a problem with it take it up to Amazon.

Next thing you know, you're whining that Amazon uses spectre_v2=off and now you suddenly want to remove that option to force their hand?

[itoffshore]

But not WinForms, which almost every single .NET/C# application I've come across uses. And no, I didn't get Wine or Mono to work with my .NET applications.

With mono + mono-basic in Arch Linux I have a VB.NET Winform web client app working (it sends messages to a Tomcat plugin)