Re:USB3: Why it's a bit harder than USB2
by MrBuddyCasino on Wednesday October 07, 2020 @06:53AM
(#24706989)
And even then, you might not want to activate the IOMMU, as the performance impact can be prohibitive in certain scenarios.
https://dl.acm.org/doi/pdf/10.1145/3230543.3230560 "For 64B DMA Reads it drops by almost 70% and even for 256B DMAs the drop is still a significant 30%".
Re:USB3: Why it's a bit harder than USB2
by segfaultbuserr on Wednesday October 07, 2020 @07:11AM
(#24707073)
At least a choice exists here. For an external port on a laptop, degraded performance is better than a disabled/unusable port, or a hacked computer. And for an internal PCI-E device, it can be turned off if risk is acceptable.
Re:USB3: Why it's a bit harder than USB2
by MrBuddyCasino on Wednesday October 07, 2020 @07:27AM
(#24707164)
Right. I think for most people, if it cannot be remotely exploited, its a non-issue and a deactivated IOMMU is the right choice. Once you really have to worry about an attack vector where your adversary has physical access, its a whole different game anyway and just flipping on the IOMMU won’t safe you.
Re:USB3: Why it's a bit harder than USB2
by josefx on Wednesday October 07, 2020 @08:28AM
(#24707495)
> Once you really have to worry about an attack vector where your adversary has physical access
There is crack the PC open physical access and just connect a device to a readily available connection access. With the later the adversary doesn't even need physical access himself, he just needs an unwary victim.
Smartphones had been plagued by something as simple as a public charging station getting compromised when Android was configured to just grant access to the file system and debug interface without requiring user confirmation. Before that Windows stopped auto playing software because it was used to infect systems from compromised devices.
[0] https://news.ycombinator.com/item?id=23991732