This week, Google Cloud paid out their highest bug bounty yet ($150k) for a vulnerability that could have been prevented with ASI [0]. Good to see that Google is pushing forward with ASI despite the performance impact, because it would benefit the security of all hosting companies that use Linux/KVM, not just the cloud providers of big tech.
My understanding was that many of the fixes for speculative execution issues themselves led to performance degradation, does anyone know the latest on that and how this compares?
Are these performance hit numbers inclusive of turning off the other mitigations?
snvzz 2 hours ago [-]
There's about one way[0] to fix timing side channels.
The RISC-V ISA has an effort to standardize a timing fence[1][2], to take care of this once and for all.
Overhead should be minimal but something is preventing it from working as well as it theoretically should. AFAIK Microsoft has been improving VBS but I don't think it's completely fixed yet.
BF6 requiring VBS (or at least "VBS capable" systems) will probably force games to find a way to deal with VBS as much as they can, but for older titles it's not always a bad idea to turn off VBS to get a less stuttery experience.
UltraSane 38 minutes ago [-]
VBS requires hyper-v to be enabled and it "owns" the CPU virtualization hardware so I can't use VMware workstation which is very annoying.
bpye 53 seconds ago [-]
VMWare Workstation [0] (and I thought VirtualBox - though I can't find any official docs [1]) should be able to use the Hyper-V hypervisor via WHP.
Same. Have to disable VBS for VirtualBox, and it gets more and more obscure with each update because some features like Windows Hello force it back on.
kookamamie 2 hours ago [-]
We're working on HPC / graphics / computer-vision software and noticed a particularly nasty issue with VBS enabled just last week. Although, have to be mentioned it was on Win10 Pro.
kachapopopow 41 minutes ago [-]
This most likely comes from IOMMU - disable it.
api 3 hours ago [-]
That's still really massive. It would only make sense in very high security environments.
Honestly running system services in VMs would be cheaper and just as good, or an OS like Qubes. VM hit is much smaller, less than 1% in some cases on newer hardware.
gpapilion 2 hours ago [-]
It makes sense in any environment you have two workloads sharing compute from two parties, public clouds.
The protection here is to ensure the vms are isolated. Without doing this there is the potential you can leak data via speculative execution across guests.
russdill 1 hours ago [-]
Look at it this way, any time a new side channel attack comes out the situation changes. Having this as a mitigation that can be turned on is helpful
eptcyka 3 hours ago [-]
VMs suffer from memory use overhead. Would be cool if the guest kernel would cooperate with the host on that.
Probably works best running VMs with the same kernel and software version.
infogulch 27 minutes ago [-]
But that just seems to reintroduce the same problem again:
> However, while KSM can reduce memory usage, it also comes with some security risks, as it can expose VMs to side-channel attacks. ...
traverseda 3 hours ago [-]
It will! For Linux hosts and Linux guests, if you use virtio and memory ballooning.
shortrounddev2 2 hours ago [-]
This was an issue for me a few years ago running docker on macOS. macOS required you to allocate memory to docker ahead of time, whereas Windows/Hyper-V was able to use memory ballooning in WSL2
api 2 hours ago [-]
It's possible to address this to some extent with ballooning memory drivers, etc.
riedel 3 hours ago [-]
From reading the article that is the exactly also the feeling of the people involved. The question is if they are on track towards e.g. the 1% eventually.
Traubenfuchs 3 hours ago [-]
Sometimes something in me starts thinking about if this regularly occurring slowing of chips through exploit mitigation is deliberate.
All of big tech wins: CPUs get slower and we need more vcpu's and more memory to serve our javascript slop to end customers: The hardware companies sell more hardware, the cloud providers sell more cloud.
gpapilion 2 hours ago [-]
I think it’s more pragmatic. We can eliminate hyperthreading to solve this, or increase memory safety at the cost of performance. One is a 50% hit in terms of vcpus, the other is now sub 50%.
Traubenfuchs 1 hours ago [-]
They also need some phony justifications though.
Can't just turn off hyperthreading.
Avamander 3 hours ago [-]
These types of mitigations have the biggest benefit when resources are shared. Do you really think cloud vendors want to lose performance to CPU or other mitigations when they could literally sell those resources to customers instead?
bzzzt 2 hours ago [-]
They don't lose anything since they sell the same instance which performs less with the mitigations on.
Customers are paying because they need more instances.
depingus 2 hours ago [-]
Sometimes its fun to engage in a little conspiratorial thinking. My 2 cents... That TPM 2.0 requirement on Windows 11 is about to create a whole ton of e-waste in October (Windows 10 EOL).
e2le 28 minutes ago [-]
I'm not so sure. Many people still ran Windows XP/7 long after the EOL date. Unless Chrome, Steam, etc drop support for Windows 10, I don't think many people will care.
bzzzt 3 hours ago [-]
Why would big tech do this when customers bring it upon themselves by building Javascript slop?
[0] https://cyberscoop.com/cloud-security-l1tf-reloaded-public-c...
Are these performance hit numbers inclusive of turning off the other mitigations?
The RISC-V ISA has an effort to standardize a timing fence[1][2], to take care of this once and for all.
0. https://tomchothia.gitlab.io/Papers/EuroSys19.pdf
1. https://lf-riscv.atlassian.net/wiki/spaces/TFXX/pages/538379...
2. https://sel4.org/Summit/2024/slides/hardware-support.pdf
What kind of workloads have noticeably lower performance with VBS?
Overhead should be minimal but something is preventing it from working as well as it theoretically should. AFAIK Microsoft has been improving VBS but I don't think it's completely fixed yet.
BF6 requiring VBS (or at least "VBS capable" systems) will probably force games to find a way to deal with VBS as much as they can, but for older titles it's not always a bad idea to turn off VBS to get a less stuttery experience.
[0] - https://techcommunity.microsoft.com/blog/virtualization/vmwa...
[1] - https://www.impostr-labs.com/use-hyper-v-and-virtualbox-toge...
Honestly running system services in VMs would be cheaper and just as good, or an OS like Qubes. VM hit is much smaller, less than 1% in some cases on newer hardware.
The protection here is to ensure the vms are isolated. Without doing this there is the potential you can leak data via speculative execution across guests.
Probably works best running VMs with the same kernel and software version.
> However, while KSM can reduce memory usage, it also comes with some security risks, as it can expose VMs to side-channel attacks. ...
All of big tech wins: CPUs get slower and we need more vcpu's and more memory to serve our javascript slop to end customers: The hardware companies sell more hardware, the cloud providers sell more cloud.
Can't just turn off hyperthreading.