Linux Lands Mitigations For Spectre-BHB / BHI On Intel & Arm, Plus An AMD Change Too

[phoronix]

Phoronix: Linux Lands Mitigations For Spectre-BHB / BHI On Intel & Arm, Plus An AMD Change Too

Within minutes of the BHI speculative execution vulnerability going public, patches were merged into the mainline Linux kernel Git tree for mitigating this offshoot from Spectre V2. The Intel and Arm processors affected by BHI (also referred to as Spectre-BHB) have mitigation work plus a change also impacts AMD processors too...

https://www.phoronix.com/scan.php?page=news_item&px=Spectre-BHI-Linux-Mitigations

[kylew77]

Are any of the *BSDs affected? I looked it up and eBPF looks like a Linux specific feature.

[CTTY]

eBPF looks like a Linux specific feature.

The vulnerability has nothing to do with eBPF. It was just used to show the vulnerability, but in theory it can be vulnerable without eBPF

Edit: see under "So, is disabling unprivileged eBPF sufficient?" [0]
[0] https://www.vusec.net/projects/bhi-spectre-bhb/

[keit99]

Are any of the *BSDs affected? I looked it up and eBPF looks like a Linux specific feature.

This vulnerability is not in the kernel interface called eBPF, but instead on a proessor level, affecting any system using the vulnerable CPUs (ARM and Intel)

[CTTY]

Do I understand it correctly, that systems saying `Full generic retpoline` in `/sys/devices/system/cpu/vulnerabilities/spectre_v2` are NOT vulnerable? (Since Full generic retpoline are applied)

[CommunityMember]

Do I understand it correctly, that systems saying `Full generic retpoline` in `/sys/devices/system/cpu/vulnerabilities/spectre_v2` are NOT vulnerable? (Since Full generic retpoline are applied)

Those CPU's already using (or forced to use) full reptolines are (mostly) considered not vulnerable. as the LFENCE and eIBRS mitigations are no longer considered sufficient.

[bezirg]

Why this sudden change of amd's spectre_v2? I thought that his new cpu vulnerability does not affect the amd side.

[asriel]

One more useless evil mitigation for one more bullshit "vulnerability". Pray that Linus will keep the possibility to switch this crap off.

[boxie]

Why this sudden change of amd's spectre_v2? I thought that his new cpu vulnerability does not affect the amd side.

AMD is vulnerable to some (not all) speculation vulns

$ grep $ /sys/devices/system/cpu/vulnerabilities/*
/sys/devices/system/cpu/vulnerabilities/itlb_multihit:Not affected
/sys/devices/system/cpu/vulnerabilities/l1tf:Not affected
/sys/devices/system/cpu/vulnerabilities/mds:Not affected
/sys/devices/system/cpu/vulnerabilities/meltdown:Not affected
/sys/devices/system/cpu/vulnerabilities/spec_store_bypass:Mitigation: Speculative Store Bypass disabled via prctl
/sys/devices/system/cpu/vulnerabilities/spectre_v1:Mitigation: usercopy/swapgs barriers and __user pointer sanitization
/sys/devices/system/cpu/vulnerabilities/spectre_v2:Mitigation: Full AMD retpoline, IBPB: conditional, STIBP: conditional, RSB filling
/sys/devices/system/cpu/vulnerabilities/srbds:Not affected
/sys/devices/system/cpu/vulnerabilities/tsx_async_abort:Not affected

[AMiguelTrindade]

One more useless evil mitigation for one more bullshit "vulnerability". Pray that Linus will keep the possibility to switch this crap off.

I have disabled all mitigations on my systems. These are "bullshit vulnerabilities" as you said. After years of being disclosed there is still to appear a working exploit.