News for hackers. Stuff that matters.

A comment on Struck by a Thunderbolt. Parent comment

Re:Struck by a Thunderbolt by the8472 on Tuesday February 26, 2019 @04:30PM (#19258464)
On the other hand IOMMU helps protecting system memory not just against malicious but also faulty devices that could otherwise write anywhere.

And people who use GPU passthrough rely on IOMMUs, so it's not like those code paths are not being exercised.

[ Parent | on HN ]
Re:Struck by a Thunderbolt by bonzini on Tuesday February 26, 2019 @05:36PM (#19258998)
Using IOMMU through the "userspace PCI" vfio driver, as done in pass-through scenarios, is a different path than using it through the DMA API.
[ Parent | on HN ]
Re:Struck by a Thunderbolt by walterbell on Wednesday February 27, 2019 @07:48AM (#19262891)
IOMMU is also exercised by Xen PCI passthrough (e.g. QubesOS network and USB driver VMs) and SR-IOV (e.g. AWS Nitro networking).
[ Parent | on HN ]
Re:Struck by a Thunderbolt by bonzini on Thursday February 28, 2019 @03:15AM (#19269852)
The problem isn't bugs in the IOMMU, it's bugs or performance issues in using the DMA API with an IOMMU.
[ Parent | on HN ]