On 5/7/2020 3:57 AM, Alex Williamson wrote:
On Mon, 4 May 2020 21:28:58 +0530
Kirti Wankhede <kwankhede@xxxxxxxxxx> wrote:
Added migration capability in IOMMU info chain.
User application should check IOMMU info chain for migration capability
to use dirty page tracking feature provided by kernel module.
Signed-off-by: Kirti Wankhede <kwankhede@xxxxxxxxxx>
---
drivers/vfio/vfio_iommu_type1.c | 15 +++++++++++++++
include/uapi/linux/vfio.h | 14 ++++++++++++++
2 files changed, 29 insertions(+)
diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
index 8b27faf1ec38..b38d278d7bff 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -2378,6 +2378,17 @@ static int vfio_iommu_iova_build_caps(struct vfio_iommu *iommu,
return ret;
}
+static int vfio_iommu_migration_build_caps(struct vfio_info_cap *caps)
+{
+ struct vfio_iommu_type1_info_cap_migration cap_mig;
+
+ cap_mig.header.id = VFIO_IOMMU_TYPE1_INFO_CAP_MIGRATION;
+ cap_mig.header.version = 1;
+ cap_mig.flags = VFIO_IOMMU_INFO_CAPS_MIGRATION_DIRTY_PAGE_TRACK;
+
+ return vfio_info_add_capability(caps, &cap_mig.header, sizeof(cap_mig));
+}
+
static long vfio_iommu_type1_ioctl(void *iommu_data,
unsigned int cmd, unsigned long arg)
{
@@ -2427,6 +2438,10 @@ static long vfio_iommu_type1_ioctl(void *iommu_data,
if (ret)
return ret;
+ ret = vfio_iommu_migration_build_caps(&caps);
+ if (ret)
+ return ret;
+
if (caps.size) {
info.flags |= VFIO_IOMMU_INFO_CAPS;
diff --git a/include/uapi/linux/vfio.h b/include/uapi/linux/vfio.h
index e3cbf8b78623..df9ce8aaafab 100644
--- a/include/uapi/linux/vfio.h
+++ b/include/uapi/linux/vfio.h
@@ -1013,6 +1013,20 @@ struct vfio_iommu_type1_info_cap_iova_range {
struct vfio_iova_range iova_ranges[];
};
+/*
+ * The migration capability allows to report supported features for migration.
+ *
+ * The structures below define version 1 of this capability.
+ */
+#define VFIO_IOMMU_TYPE1_INFO_CAP_MIGRATION 1
+
+struct vfio_iommu_type1_info_cap_migration {
+ struct vfio_info_cap_header header;
+ __u32 flags;
+ /* supports dirty page tracking */
+#define VFIO_IOMMU_INFO_CAPS_MIGRATION_DIRTY_PAGE_TRACK (1 << 0)
+};
+
What about exposing the maximum supported dirty bitmap size and the
supported page sizes? Thanks,
How should user application use that?
Thanks,
Kirti
Alex
#define VFIO_IOMMU_GET_INFO _IO(VFIO_TYPE, VFIO_BASE + 12)
/**
