On Wed, 28 Feb 2024 at 15:17, Zhenzhong Duan <zhenzhong.d...@intel.com> wrote:
> When there is VFIO device and vIOMMU cap/ecap is updated based on host
* cap/ecap -> capability/extended capability registers are updated ...
> IOMMU cap/ecap, migration should be blocked.
* It'll help to mention why migration should be blocked in this case?
> Signed-off-by: Zhenzhong Duan <zhenzhong.d...@intel.com>
> ---
> hw/i386/intel_iommu.c | 16 ++++++++++++++--
> +static Error *vtd_mig_blocker;
> +
> static int vtd_check_iommufd_hdev(IntelIOMMUState *s,
> IOMMUFDDevice *idev,
> Error **errp)
> @@ -3861,8 +3864,17 @@ static int vtd_check_iommufd_hdev(IntelIOMMUState *s,
> tmp_cap |= VTD_CAP_MGAW(host_mgaw + 1);
> }
>
> - s->cap = tmp_cap;
> - return 0;
> + if (s->cap != tmp_cap) {
> + if (vtd_mig_blocker == NULL) {
> + error_setg(&vtd_mig_blocker,
> + "cap/ecap update from host IOMMU block migration");
> + ret = migrate_add_blocker(&vtd_mig_blocker, errp);
> + }
> + if (!ret) {
> + s->cap = tmp_cap;
> + }
> + }
> + return ret;
* I couldn't find vtd_check_* function in the tree, but what happens
if vtd_mig_blocker != NULL? What will be 'ret' then?
Thank you.
---
- Prasad
