Spectre mitigations murder userspace performance

Re:Spectre mitigations murder userspace performance by reader_mode on Saturday June 19, 2021 @09:17AM (#27560206)

How long would you have to run JavaScript on your desktop to leak sensitive information and has there been any known exploits in the wild ?

[ Parent | on HN ]

Re:Spectre mitigations murder userspace performance by mhh__ on Saturday June 19, 2021 @09:41AM (#27560384)

https://leaky.page/

I don't know about the wild, but if this were tuned (i.e. this requires a lot of work for the first byte, the rest are easy) for a HVT you wouldn't know.

[ Parent | on HN ]

Re:Spectre mitigations murder userspace performance by ggreer on Saturday June 19, 2021 @01:53PM (#27562656)

On my laptop's Core i5-10210U with mitigations=off, the demo just prints "[!] error: could not infer memory layout" until it runs out of retries.

This is on Chromium 91.0.4472.106 and kernel 5.12.11-arch1-1. lscpu shows vulnerabilities:

    Itlb multihit:         KVM: Mitigation: VMX disabled
    L1tf:                  Not affected
    Mds:                   Not affected
    Meltdown:              Not affected
    Spec store bypass:     Vulnerable
    Spectre v1:            Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers
    Spectre v2:            Vulnerable, IBPB: disabled, STIBP: disabled
    Srbds:                 Mitigation; TSX disabled
    Tsx async abort:       Not affected

[ Parent | on HN ]

Re:Spectre mitigations murder userspace performance by diveanon on Saturday June 19, 2021 @07:40PM (#27565591)

I got the same result on a pretty similar system.

[ Parent | on HN ]

Re:Spectre mitigations murder userspace performance by magila on Saturday June 19, 2021 @11:39AM (#27561426)

The problem is that PoC is extracting data which the PoC itself created specifically to facilitate said extraction. AFAIK no one has created a PoC which can extract specific data which hasn't been constructed to facilitate the PoC.

[ Parent | on HN ]