Re: [PATCH 1/1] amd_iommu: fix device entry invalidation

[Philippe Mathieu-Daudé]

On 6/1/21 4:10 PM, Roman Kapl wrote:
> Since QEMU maintains TLB entries keyed by the device ID, there is no separate
> device entry cache. This means we need to invalidate all translations keyed by
> the device ID.
> 

Fixes: d29a09ca684 ("hw/i386: Introduce AMD IOMMU") ?

> Signed-off-by: Roman Kapl <rka@sysgo.com>
> ---
>  hw/i386/amd_iommu.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c
> index 2801dff97c..62d22b220b 100644
> --- a/hw/i386/amd_iommu.c
> +++ b/hw/i386/amd_iommu.c
> @@ -356,6 +356,7 @@ static void amdvi_update_iotlb(AMDVIState *s, uint16_t 
> devid,
>          }
>  
>          entry->domid = domid;
> +        entry->devid = devid;
>          entry->perms = to_cache.perm;
>          entry->translated_addr = to_cache.translated_addr;
>          entry->page_mask = to_cache.addr_mask;
> @@ -399,6 +400,8 @@ static void amdvi_inval_devtab_entry(AMDVIState *s, 
> uint64_t *cmd)
>          amdvi_log_illegalcom_error(s, extract64(cmd[0], 60, 4),
>                                     s->cmdbuf + s->cmdbuf_head);
>      }
> +    g_hash_table_foreach_remove(s->iotlb, amdvi_iotlb_remove_by_devid,
> +                                &devid);
>      trace_amdvi_devtab_inval(PCI_BUS_NUM(devid), PCI_SLOT(devid),
>                               PCI_FUNC(devid));
>  }
>