Re: [PATCH 3/5] vDPA: introduce vDPA bus
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: 2020-01-20 12:09:22
Also in:
kvm, lkml, virtualization
Possibly related (same subject, not in this thread)
- 2020-01-21 · Re: [PATCH 3/5] vDPA: introduce vDPA bus · Jason Gunthorpe <hidden>
- 2020-01-21 · Re: [PATCH 3/5] vDPA: introduce vDPA bus · "Michael S. Tsirkin" <mst@redhat.com>
- 2020-01-21 · Re: [PATCH 3/5] vDPA: introduce vDPA bus · Jason Gunthorpe <hidden>
- 2020-01-21 · Re: [PATCH 3/5] vDPA: introduce vDPA bus · Jason Wang <jasowang@redhat.com>
- 2020-01-21 · RE: [PATCH 3/5] vDPA: introduce vDPA bus · "Tian, Kevin" <kevin.tian@intel.com>
On Mon, Jan 20, 2020 at 04:44:34PM +0800, Jason Wang wrote:
On 2020/1/19 下午5:59, Michael S. Tsirkin wrote:quoted
On Sun, Jan 19, 2020 at 09:07:09AM +0000, Shahaf Shuler wrote:quoted
quoted
Technically, we can keep the incremental API here and let the vendor vDPA drivers to record the full mapping internally which may slightly increase the complexity of vendor driver.What will be the trigger for the driver to know it received the last mapping on this series and it can now push it to the on-chip IOMMU?Some kind of invalidate API?The problem is how to deal with the case of vIOMMU. When vIOMMU is enabling there's no concept of last mapping. Thanks
Most IOMMUs have a translation cache so have an invalidate API too. -- MST