On 9/21/2024 1:56 AM, Alejandro Jimenez wrote:
> In subject:
> s/invaldate/invalidate/
>
> On 9/16/24 10:31, Santosh Shukla wrote:
>> From: Suravee Suthikulpanit <suravee.suthikulpa...@amd.com>
>>
>> In order to support AMD IOMMU interrupt remapping emulation with PCI
>> pass-through devices, QEMU needs to notify VFIO when guest IOMMU driver
>> updates and invalidate the guest interrupt remapping table (IRT), and
>> communicate information so that the host IOMMU driver can update
>> the shadowed interrupt remapping table in the host IOMMU.
>>
>> Therefore, send notification when guet
>
> s/guet/guest
>
> Alejandro
V3. Thanks
>
> IOMMU emulates the IRT invalidation
>> commands.
>>
>> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpa...@amd.com>
>> Signed-off-by: Santosh Shukla <santosh.shu...@amd.com>
>> ---
>> hw/i386/amd_iommu.c | 12 ++++++++++++
>> 1 file changed, 12 insertions(+)
>>
>> diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c
>> index 24fcd561345c..9095146525e6 100644
>> --- a/hw/i386/amd_iommu.c
>> +++ b/hw/i386/amd_iommu.c
>> @@ -431,6 +431,12 @@ static void amdvi_complete_ppr(AMDVIState *s, uint64_t
>> *cmd)
>> trace_amdvi_ppr_exec();
>> }
>> +static void amdvi_intremap_inval_notify_all(AMDVIState *s, bool global,
>> + uint32_t index, uint32_t mask)
>> +{
>> + x86_iommu_iec_notify_all(X86_IOMMU_DEVICE(s), global, index, mask);
>> +}
>> +
>> static void amdvi_inval_all(AMDVIState *s, uint64_t *cmd)
>> {
>> if (extract64(cmd[0], 0, 60) || cmd[1]) {
>> @@ -438,6 +444,9 @@ static void amdvi_inval_all(AMDVIState *s, uint64_t *cmd)
>> s->cmdbuf + s->cmdbuf_head);
>> }
>> + /* Notify global invalidation */
>> + amdvi_intremap_inval_notify_all(s, true, 0, 0);
>> +
>> amdvi_iotlb_reset(s);
>> trace_amdvi_all_inval();
>> }
>> @@ -486,6 +495,9 @@ static void amdvi_inval_inttable(AMDVIState *s, uint64_t
>> *cmd)
>> return;
>> }
>> + /* Notify global invalidation */
>> + amdvi_intremap_inval_notify_all(s, true, 0, 0);
>> +
>> trace_amdvi_intr_inval();
>> }
>>
