---
name: "criu-amdgpu-plugin"
section: "1"
platform: "linux"
package: "criu"
version: "4.2.1-1"
release: "Arch Linux 2026-11-01"
platforms: ["linux"]
url: "https://new.carta.tech/criu-amdgpu-plugin"
---

# criu-amdgpu-plugin(1)

." make the size of the head bigger **NAME** **criu-amdgpu-plugin - A plugin extension to CRIU to support checkpoint/restore in userspace for AMD GPUs.**

**." make the size of the head bigger** **CURRENT SUPPORT**

**Single and Multi GPU systems (Gfx9) Checkpoint / Restore on different system Checkpoint / Restore inside a docker container Pytorch Tensorflow Using CRIU Image Streamer**

**." make the size of the head bigger** **DESCRIPTION**

**Though criu** is a great tool for checkpointing and restoring running applications, it has certain limitations such as it cannot handle applications that have device files open. In order to support **ROCm** based workloads with **criu** we need to augment criu’s core functionality with a plugin based extension mechanism. **criu-amdgpu-plugin** provides the necessary support to criu to allow Checkpoint / Restore with ROCm.

." make the size of the head bigger **DEPENDENCIES**

**amdkfd support**

In order to snapshot the **VRAM** and other **GPU** device states, we require an updated version of amdkfd(amdgpu) driver.

." make the size of the head bigger **OPTIONS**

**Optional parameters can be passed in as environment variables before executing criu command.**

**KFD\_FW\_VER\_CHECK**

Enable or disable firmware version check. If enabled, firmware version on restored gpu needs to be greater than or equal firmware version on checkpointed GPU. Default:Enabled

```
E.g:
KFD_FW_VER_CHECK=0
```

**KFD\_SDMA\_FW\_VER\_CHECK**

Enable or disable SDMA firmware version check. If enabled, SDMA firmware version on restored gpu needs to be greater than or equal firmware version on checkpointed GPU. Default:Enabled

```
E.g:
KFD_SDMA_FW_VER_CHECK=0
```

**KFD\_CACHES\_COUNT\_CHECK**

Enable or disable caches count check. If enabled, the caches count on restored GPU needs to be greater than or equal caches count on checkpointed GPU. Default:Enabled

```
E.g:
KFD_CACHES_COUNT_CHECK=0
```

**KFD\_NUM\_GWS\_CHECK**

Enable or disable num\_gws check. If enabled, the num\_gws on restored GPU needs to be greater than or equal num\_gws on checkpointed GPU. Default:Enabled

```
E.g:
KFD_NUM_GWS_CHECK=0
```

**KFD\_VRAM\_SIZE\_CHECK**

Enable or disable VRAM size check. If enabled, the VRAM size on restored GPU needs to be greater than or equal VRAM size on checkpointed GPU. Default:Enabled

```
E.g:
KFD_VRAM_SIZE_CHECK=0
```

**KFD\_NUMA\_CHECK**

Enable or disable NUMA CPU region check. If enabled, the plugin will restore GPUs that belong to one CPU NUMA region to the same CPU NUMA region. Default:Enabled

```
E.g:
KFD_NUMA_CHECK=1
```

**KFD\_CAPABILITY\_CHECK**

Enable or disable capability check. If enabled, the capability on restored GPU needs to be equal to the capability on the checkpointed GPU. Default:Enabled

```
E.g:
KFD_CAPABILITY_CHECK=1
```

**KFD\_MAX\_BUFFER\_SIZE**

On some systems, VRAM sizes may exceed RAM sizes, and so buffers for dumping and restoring VRAM may be unable to fit. Set to a nonzero value (in bytes) to set a limit on the plugin’s memory usage. Default:0 (Disabled)

```
E.g:
KFD_MAX_BUFFER_SIZE="2G"
```

." make the size of the head bigger **AUTHOR**

**The AMDKFD team.**

**." make the size of the head bigger** **COPYRIGHT**

**Copyright (C) 2020-2021, Advanced Micro Devices, Inc. (AMD)**