kata-types: Introduce GPU related fields in RemoteInfo

To provide the remote hypervisor with the necessary intelligence
to select the most appropriate instance for a given GPU instance,
leading to better resource allocation, two fields `default_gpus`
and `default_gpu_model` are introduced in `RemoteInfo`.

Fixes #10484

Signed-off-by: alex.lyn <alex.lyn@antgroup.com>
This commit is contained in:
alex.lyn 2025-06-26 15:04:35 +08:00
parent 648fef4f52
commit ed0c0b2367

View File

@ -1110,6 +1110,14 @@ pub struct RemoteInfo {
/// Remote hyperisor timeout of creating (in seconds) /// Remote hyperisor timeout of creating (in seconds)
#[serde(default)] #[serde(default)]
pub hypervisor_timeout: i32, pub hypervisor_timeout: i32,
/// GPU specific annotations (currently only applicable for Remote Hypervisor)
/// default_gpus specifies the number of GPUs required for the Kata VM
#[serde(default)]
pub default_gpus: u32,
/// default_gpu_model specifies GPU model like tesla, h100, a100, readeon etc.
#[serde(default)]
pub default_gpu_model: String,
} }
/// Common configuration information for hypervisors. /// Common configuration information for hypervisors.