Microsoft Proposes "Hornet" Security Module For The Linux Kernel
phoronix.com/news/Microsoft-Ho…
Microsoft Proposes "Hornet" Security Module For The Linux Kernel
Microsoft's newest open-source contribution to the Linux kernel being proposed is..www.phoronix.com
This entry was edited (2 days ago)
like this
Trent
in reply to ikidd • • •like this
Ⓜ3️⃣3️⃣ 🌌, TVA and Ⓜ3️⃣3️⃣ 🌌 like this.
cronenthal
in reply to ikidd • • •like this
Ⓜ3️⃣3️⃣ 🌌, TVA and Ⓜ3️⃣3️⃣ 🌌 like this.
Daniel Quinn
in reply to ikidd • • •like this
Ⓜ3️⃣3️⃣ 🌌 and TVA like this.
Rin
in reply to Daniel Quinn • • •Leadership | Linux Foundation
www.linuxfoundation.orglike this
Ⓜ3️⃣3️⃣ 🌌 likes this.
Phoenixz
in reply to Rin • • •grue
in reply to Phoenixz • • •Rin
in reply to grue • • •MonkderVierte
in reply to Rin • • •That it is.
Norah (pup/it/she)
in reply to Rin • • •3aqn5k6ryk
in reply to ikidd • • •like this
Ⓜ3️⃣3️⃣ 🌌 and TVA like this.
macaw_dean_settle
in reply to 3aqn5k6ryk • • •3aqn5k6ryk
in reply to macaw_dean_settle • • •spittingimage
in reply to ikidd • • •like this
Inf_V and TVA like this.
caseyweederman
in reply to ikidd • • •Ah yes, the "extended Berkeley Packet Filter".
Wikipedia:
Phoronix:
... show moreSo this is to make kernel-level instructions from userspace (somethi
Ah yes, the "extended Berkeley Packet Filter".
Wikipedia:
Phoronix:
So this is to make kernel-level instructions from userspace (something that's already happening) more secure.
The thread linked by the OP is Jarkko Sakkinen (kernel maintainer) seemingly saying "show your work, your patch is full of nonsense" in a patch submitted for review to the Linux kernel.
Edit: the OP has edited the link, it used to point to this comment in the mailing list chain.
Microsoft Proposes "Hornet" Security Module For The Linux Kernel
www.phoronix.comlike this
TVA likes this.
interdimensionalmeme
in reply to caseyweederman • • •Magiilaro
in reply to caseyweederman • • •Loading BPF code from user space is, I hope, only possible with root access to the system.
That would mean that an attacker needs root access to exploit BPF, but if an attacker has root access what stops him/her to do anything they want? At this time the system is lost anyway.
Or am I missing anything?
ozymandias117
in reply to Magiilaro • • •If the executable binary has to be signed with a key, similar to the module signing key, Microsoft could sign their binaries
This, along with secureboot, would prevent the owner of the machine from running eBPF programs Microsoft doesn't want you to run, even with root
Magiilaro
in reply to ozymandias117 • • •lumony
in reply to Magiilaro • • •Odds are because there isn't one.
Abusers will always try to justify their abuse by saying their victims "don't understand" why it's "necessary."
ozymandias117
in reply to Magiilaro • • •Magiilaro
in reply to ozymandias117 • • •mina86
in reply to caseyweederman • • •That’s not what he’s saying. He’s saying: ‘You’re using terms which aren’t that familiar to everyone. Could you explain them?’
EatMyPixelDust
in reply to ikidd • • •like this
TVA likes this.
Magiilaro
in reply to ikidd • • •Matt
in reply to Magiilaro • • •Magiilaro
in reply to Matt • • •Yes I can.
But I am a Linux system administrator with 20 years of experience.
This should not be the level of measurement for stuff like this. 😉
What I meant was: Don't put a Microsoft master trusted authority in the Kernel, unless one chooses to install a Microsoft distribution.
And don't go the SSL/TLS way with the huge number of default authorities that get installed on every system.
It would be a pain to be forced to always build my own Kernel again just to keep Microsoft or any other institution/company that I find untrustworthy out of it.
like this
Ⓜ3️⃣3️⃣ 🌌 likes this.
Ⓜ3️⃣3️⃣ 🌌
in reply to ikidd • •Linux reshared this.
ikidd
in reply to Ⓜ3️⃣3️⃣ 🌌 • • •BaconIsAVeg
in reply to ikidd • • •VSCode is one of the best free editors second only to Neovim (and maybe DoomEmacs), and the world runs off GitHub whether we like it or not. Azure runs Linux, and a lot of work has been put into WSL to where it's pretty darn handy if you're forced to use company Windows hardware but need to do Dev/SRE tasks.
Windows 11 and Teams though can die in a tire fire.
ikidd
in reply to BaconIsAVeg • • •If you don't want telemetry, you have to use VScodium, and then you don't get to use marketplace. Github didn't start as a Microsoft project or it would be far more enshittified than it is now, but even so Microsoft is sure trying to fuck that up with their Copilot bullshit.
WSL is the definitition of EEE, and has prevented a great deal of Linux-ward movement that might have happened without it, even with IT department resistance. It's a crutch to keep devs from having to go to Linux to get the useful tools, like docker which is a mess on Windows, but just usable enough to get by.
And oh, yes, Teams can get shot with a ball of its own shit and fall into the dumpster fire.
like this
Ⓜ3️⃣3️⃣ 🌌 likes this.
Ⓜ3️⃣3️⃣ 🌌
in reply to ikidd • •Linux reshared this.
utopiah
in reply to BaconIsAVeg • • •It doesn't and we don't like it anyway.
PS: to clarify, yes GitHub is wildly popular but, and the kernel is a particularly interesting example, it does not host ALL projects, only a lot of popular ones. A lot of very popular ones are also NOT there but rather on their own git, mailing list, GitLab instance, Gitea, etc. It's a shortcut, I understand that, but by ascertaining it as "truth" it's hiding a reality that is quite different and showing that reliable alternatives do exist.
WasPentalive
in reply to ikidd • • •They probably named it HORNET for a reason - think Japanese Murder Hornets... What Could Possibly Go Wrong??
It will probably start out as little glitches and slowdowns to destroy faith in your system ("Windows works right all the time") a random 2 second pauses. Finally one day every Linux box in the world crashes, all at the same time, because some 'dummy' in Microsoft deleted the private signing key.
like this
Ⓜ3️⃣3️⃣ 🌌 likes this.
Ⓜ3️⃣3️⃣ 🌌
in reply to WasPentalive • •Linux reshared this.
lumony
in reply to ikidd • • •InnerScientist
in reply to lumony • • •Preventing kernel modifications to expand upon the work done for kernel lockdown. Add additional layers to system security.
Kernel_lockdown:
lumony
in reply to InnerScientist • • •InnerScientist
in reply to lumony • • •From the mailing list I'm reading that kernel maintainers have heard a few companies looking for something like this, so yes?
Edit:
priapus
in reply to ikidd • • •Do people in this thread not understand that Microsoft frequently contributes to Linux? They've already lost the battle there. They rely on Linux for servers as much as everybody else.
Not necessarily saying this is a good thing or not, but writing off any Linux contributions Microsoft makes would be pretty silly.
FauxLiving
in reply to priapus • • •Their contributions are welcome and appreciated.
But, given Microsoft's history, any suggestions from them should be treated with skepticism.
alphadont
in reply to FauxLiving • • •It's not like it's a proprietary blob. No one is stupid enough to accept a proprietary security blob from Microsoft.
Moreover, if you click through to the article, you see that this module entirely concerns eBPF, which is essentially unused outside of corporate servers (and Android phones) in the first place and is therefore barely our business to begin with.