2024-10-02 14:42:34 -04:00
---
myst:
html_meta:
"description lang=en": "Learn how to use the AMD SMI command line tool."
"keywords": "api, smi, lib, system, management, interface, example"
---
# AMD SMI CLI tool usage
This tool is a command line interface (CLI) for manipulating and monitoring the
`amdgpu` kernel; it is intended to replace and deprecate the existing `rocm_smi`
CLI tool and `gpuv-smi` tool. The AMD SMI CLI tool uses Ctypes to call the
`amd_smi_lib` API.
When using the CLI tool, you should have at least one AMD GPU and the driver
installed.
```{admonition} Disclaimer
The AMD SMI CLI tool is provided as an example code to aid the development of
telemetry tools. The [Python](./amdsmi-py-lib) or [C++
library](./amdsmi-cpp-lib) is recommended as a robust data source.
` ``
## Install the CLI Tool and Python library
Refer to the [installation instructions](../install/install.md).
2024-04-15 10:16:39 -07:00
2024-10-02 14:42:34 -04:00
## Get started
2024-08-09 13:07:02 -05:00
2025-05-14 12:43:33 -04:00
The ` amd-smi` command provides system management and monitoring capabilities for
2024-10-02 14:42:34 -04:00
AMD hardware. When run without arguments, it reports the version and platform
detected:
2024-04-15 10:16:39 -07:00
2024-10-02 14:42:34 -04:00
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi
usage: amd-smi [-h] ...
2025-12-15 13:19:30 -06:00
AMD System Management Interface | Version: 26.2.1 | ROCm version: 7.2.0 | Platform: Linux Baremetal
2024-04-15 10:16:39 -07:00
options:
-h, --help show this help message and exit
AMD-SMI Commands:
Descriptions:
version Display version information
list List GPU information
static Gets static information about the specified GPU
firmware (ucode) Gets firmware information about the specified GPU
bad-pages Gets bad page information about the specified GPU
metric Gets metric/performance information about the specified GPU
2025-09-26 17:33:01 -05:00
process Lists compute process information running on the specified GPU
2024-04-15 10:16:39 -07:00
event Displays event information for the given GPU
topology Displays topology information of the devices
set Set options for devices
reset Reset options for devices
2024-10-02 14:42:34 -04:00
monitor (dmon) Monitor metrics for target devices
2024-04-15 10:16:39 -07:00
xgmi Displays xgmi information of the devices
2025-03-13 05:59:34 -04:00
partition Displays partition information of the devices
2025-05-30 20:08:09 -05:00
ras Retrieve RAS (CPER) entries from the driver
2024-04-15 10:16:39 -07:00
` ``
2024-09-03 15:57:39 -05:00
Example commands:
2024-10-02 14:42:34 -04:00
` ``shell-session
2024-09-03 15:57:39 -05:00
amd-smi static --gpu 0
amd-smi metric
amd-smi process --gpu 0 1
amd-smi reset --gpureset --gpu all
` ``
2024-10-02 14:42:34 -04:00
` ``{note}
For command-specific help, use ` amd-smi [command] --help` for see more detailed
usage information. See [Commands](#cmds).
2024-04-15 10:16:39 -07:00
2024-10-02 14:42:34 -04:00
For more detailed version information, use ` amd-smi version`.
` ``
2024-04-15 10:16:39 -07:00
2025-08-05 19:58:37 -05:00
Environment variables:
You can set one or more variables in front of any ` amd-smi` invocation. For example:
` ``shell-session
AMDSMI_GPU_METRICS_CACHE_MS=200 amd-smi metric
` ``
Current Variables:
` ``{note}
AMDSMI_GPU_METRICS_CACHE_MS - Controls the internal GPU metrics cache duration (ms). Default 100, set to 0 to disable.
2025-10-08 21:48:08 -05:00
AMDSMI_ASIC_INFO_CACHE_MS - Controls the internal GPU asic info cache duration (ms). Default 10000, set to 0 to disable.
2025-08-05 19:58:37 -05:00
` ``
2024-10-02 14:42:34 -04:00
(cmds)=
2024-04-15 10:16:39 -07:00
## Commands
2024-10-02 14:42:34 -04:00
The following are the help output for each command, providing quick reference
details for usage.
2024-04-15 10:16:39 -07:00
2024-10-02 14:42:34 -04:00
(cmd-list)=
### amd-smi list
Lists GPU information.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi list --help
usage: amd-smi list [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]]
2025-05-30 20:08:09 -05:00
Lists all detected devices on the system.
Lists the BDF, UUID, KFD_ID, NODE_ID, and Partition ID for each GPU and/or CPUs.
2024-06-18 23:09:01 -05:00
In virtualization environments, it can also list VFs associated to each
GPU with some basic information for each VF.
2024-04-15 10:16:39 -07:00
2025-05-30 20:08:09 -05:00
List Arguments:
-h, --help show this help message and exit
-e Enumeration mapping to other features.
Includes CARD, RENDER, HSA_ID, HIP_ID, and HIP_UUID.
Device Arguments:
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
2024-04-15 10:16:39 -07:00
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
2025-05-30 20:08:09 -05:00
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
2024-04-15 10:16:39 -07:00
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-static)=
### amd-smi static
Gets static information about the specified GPU. See the [sample
output](#cli-ex-static) for ` amd-smi static`.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi static --help
2024-06-18 23:09:01 -05:00
usage: amd-smi static [-h] [-g GPU [GPU ...] | -U CPU [CPU ...]] [-a] [-b] [-V] [-d] [-v]
[-c] [-B] [-R] [-r] [-p] [-l] [-P] [-x] [-u] [-s] [-i]
[--json | --csv] [--file FILE] [--loglevel LEVEL]
2024-04-15 10:16:39 -07:00
2024-06-18 23:09:01 -05:00
If no GPU is specified, returns static information for all GPUs on the system.
If no static argument is provided, all static information will be displayed.
2024-04-15 10:16:39 -07:00
Static Arguments:
-h, --help show this help message and exit
-a, --asic All asic information
-b, --bus All bus information
2025-09-16 19:51:13 -05:00
-I, --ifwi All video bios\IFWI information (if available)
2024-04-15 10:16:39 -07:00
-d, --driver Displays driver version
-v, --vram All vram information
-c, --cache All cache information
-B, --board All board information
2024-06-18 23:09:01 -05:00
-R, --process-isolation The process isolation status
2025-05-30 20:08:09 -05:00
-r, --ras Displays RAS features information;
Sudo may be required for some features
2024-12-04 00:10:32 -06:00
-C, --clock [CLOCK ...] Show one or more valid clock frequency levels. Available options:
SYS, DF, DCEF, SOC, MEM, VCLK0, VCLK1, DCLK0, DCLK1, ALL
2024-04-15 10:16:39 -07:00
-p, --partition Partition information
-l, --limit All limit metric values (i.e. power and thermal limits)
2025-02-18 15:04:17 -06:00
-P, --soc-pstate The available soc pstate policy
2024-06-18 23:09:01 -05:00
-x, --xgmi-plpd The available XGMI per-link power down policy
2024-04-15 10:16:39 -07:00
-u, --numa All numa node information
CPU Arguments:
-s, --smu All SMU FW information
-i, --interface-ver Displays hsmp interface version
2025-05-30 20:08:09 -05:00
Device Arguments:
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
2024-04-15 10:16:39 -07:00
Command Modifiers:
2025-05-30 20:08:09 -05:00
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
2024-04-15 10:16:39 -07:00
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-firmware)=
### amd-smi firmware
Gets firmware information about the specified GPU.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi firmware --help
usage: amd-smi firmware [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]] [-f]
If no GPU is specified, return firmware information for all GPUs on the system.
Firmware Arguments:
-h, --help show this help message and exit
2025-05-30 20:08:09 -05:00
-f, --ucode-list, --fw-list All FW list information
Device Arguments:
2024-04-15 10:16:39 -07:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-04-15 10:16:39 -07:00
all | Selects all devices
2025-05-30 20:08:09 -05:00
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
2024-04-15 10:16:39 -07:00
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-bad-pages)=
### amd-smi bad-pages
Gets bad page information about the specified GPU.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi bad-pages --help
usage: amd-smi bad-pages [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]] [-p]
[-r] [-u]
If no GPU is specified, return bad page information for all GPUs on the system.
Bad Pages Arguments:
2025-05-30 20:08:09 -05:00
-h, --help show this help message and exit
-p, --pending Displays all pending retired pages
-r, --retired Displays retired pages
-u, --un-res Displays unreservable pages
Device Arguments:
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
2024-04-15 10:16:39 -07:00
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-metric)=
### amd-smi metric
Gets metrics and performance information about the specified GPU.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi metric --help
usage: amd-smi metric [-h] [-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]]
[-w INTERVAL] [-W TIME] [-i ITERATIONS] [-m] [-u] [-p] [-c] [-t]
[-P] [-e] [-k] [-f] [-C] [-o] [-l] [-x] [-E] [--cpu-power-metrics]
[--cpu-prochot] [--cpu-freq-metrics] [--cpu-c0-res]
2025-01-31 08:11:51 -06:00
[--cpu-lclk-dpm-level NBIOID] [--cpu-pwr-svi-telemetry-rails]
2024-04-15 10:16:39 -07:00
[--cpu-io-bandwidth IO_BW LINKID_NAME]
[--cpu-xgmi-bandwidth XGMI_BW LINKID_NAME] [--cpu-metrics-ver]
[--cpu-metrics-table] [--cpu-socket-energy] [--cpu-ddr-bandwidth]
[--cpu-temp] [--cpu-dimm-temp-range-rate DIMM_ADDR]
[--cpu-dimm-pow-consumption DIMM_ADDR]
[--cpu-dimm-thermal-sensor DIMM_ADDR] [--core-boost-limit]
[--core-curr-active-freq-core-limit] [--core-energy]
[--json | --csv] [--file FILE] [--loglevel LEVEL]
2024-06-18 23:09:01 -05:00
If no GPU is specified, returns metric information for all GPUs on the system.
2024-11-11 16:07:11 -06:00
If no metric argument is provided, all metric information will be displayed.
2024-04-15 10:16:39 -07:00
Metric arguments:
2025-05-30 20:08:09 -05:00
-h, --help show this help message and exit
-m, --mem-usage Memory usage per block
-u, --usage Displays engine usage information
-p, --power Current power usage
-c, --clock Average, max, and current clock frequencies
-t, --temperature Current temperatures
-P, --pcie Current PCIe speed, width, and replay count
-e, --ecc Total number of ECC errors
-k, --ecc-blocks Number of ECC errors per block
-V, --voltage GPU voltage
-f, --fan Current fan speed
-C, --voltage-curve Display voltage curve
-o, --overdrive Current GFX and MEM clock overdrive level
-l, --perf-level Current DPM performance level
-x, --xgmi-err XGMI error information since last read
-E, --energy Amount of energy consumed
2025-07-09 16:38:09 -05:00
-v, --violation Displays throttle accumulators;
2025-05-30 20:08:09 -05:00
Only available for MI300 or newer ASICs
Watch Arguments:
-w, --watch INTERVAL Reprint the command in a loop of INTERVAL seconds
-W, --watch_time TIME The total duration of TIME to watch the command
-i, --iterations ITERATIONS The total number of ITERATIONS to repeat the command
2024-04-15 10:16:39 -07:00
CPU Arguments:
--cpu-power-metrics CPU power metrics
--cpu-prochot Displays prochot status
--cpu-freq-metrics Displays currentFclkMemclk frequencies and cclk frequency limit
--cpu-c0-res Displays C0 residency
--cpu-lclk-dpm-level NBIOID Displays lclk dpm level range. Requires socket ID and NBOID as inputs
2025-02-18 15:04:17 -06:00
--cpu-pwr-svi-telemetry-rails Displays svi based telemetry for all rails
2025-12-15 13:20:47 -06:00
--cpu-io-bandwidth IO_BW LINKID_NAME Displays current IO bandwidth for the selected CPU.
input parameters are bandwidth type(1) and link ID encodings
2024-04-15 10:16:39 -07:00
i.e. P2, P3, G0 - G7
2025-12-15 13:20:47 -06:00
--cpu-xgmi-bandwidth XGMI_BW LINKID_NAME Displays current XGMI bandwidth for the selected CPU
input parameters are bandwidth type(1,2,4) and link ID encodings
2024-04-15 10:16:39 -07:00
i.e. P2, P3, G0 - G7
--cpu-metrics-ver Displays metrics table version
--cpu-metrics-table Displays metric table
--cpu-socket-energy Displays socket energy for the selected CPU socket
2025-12-15 13:20:47 -06:00
--cpu-ddr-bandwidth Displays per socket max ddr bw, current utilized bw,
2024-04-15 10:16:39 -07:00
and current utilized ddr bw in percentage
--cpu-temp Displays cpu socket temperature
--cpu-dimm-temp-range-rate DIMM_ADDR Displays dimm temperature range and refresh rate
--cpu-dimm-pow-consumption DIMM_ADDR Displays dimm power consumption
--cpu-dimm-thermal-sensor DIMM_ADDR Displays dimm thermal sensor
2026-01-06 10:37:07 -06:00
--cpu-dfcstate-ctrl Displays DFCState control status
--cpu-railisofreq-policy Displays CPU ISO frequency policy
2024-04-15 10:16:39 -07:00
CPU Core Arguments:
--core-boost-limit Get boost limit for the selected cores
--core-curr-active-freq-core-limit Get Current CCLK limit set per Core
--core-energy Displays core energy for the selected core
2025-05-30 20:08:09 -05:00
Device Arguments:
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
2024-04-15 10:16:39 -07:00
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-process)=
### amd-smi process
2025-09-26 17:33:01 -05:00
Lists compute process information running on the specified GPU.
2024-10-02 14:42:34 -04:00
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi process --help
usage: amd-smi process [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]]
[-w INTERVAL] [-W TIME] [-i ITERATIONS] [-G] [-e] [-p PID]
[-n NAME]
2024-06-18 23:09:01 -05:00
If no GPU is specified, returns information for all GPUs on the system.
2024-11-11 16:07:11 -06:00
If no process argument is provided, all process information will be displayed.
2024-04-15 10:16:39 -07:00
2025-05-30 20:08:09 -05:00
Process arguments:
2024-04-15 10:16:39 -07:00
Process arguments:
-h, --help show this help message and exit
2025-05-30 20:08:09 -05:00
-G, --general pid, process name, memory usage
-e, --engine All engine usages
-p, --pid PID Gets all process information about the specified process based on Process ID
-n, --name NAME Gets all process information about the specified process based on Process Name.
If multiple processes have the same name, information is returned for all of them.
Watch Arguments:
-w, --watch INTERVAL Reprint the command in a loop of INTERVAL seconds
-W, --watch_time TIME The total duration of TIME to watch the command
-i, --iterations ITERATIONS The total number of ITERATIONS to repeat the command
Device Arguments:
2024-04-15 10:16:39 -07:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-04-15 10:16:39 -07:00
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-event)=
### amd-smi event
Displays event information for the given GPU.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi event --help
usage: amd-smi event [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]]
If no GPU is specified, returns event information for all GPUs on the system.
Event Arguments:
-h, --help show this help message and exit
2025-05-30 20:08:09 -05:00
Device Arguments:
2024-04-15 10:16:39 -07:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-04-15 10:16:39 -07:00
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-topology)=
### amd-smi topology
Displays topology information of the specified devices.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi topology --help
usage: amd-smi topology [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]] [-a]
[-w] [-o] [-t] [-b]
2024-06-18 23:09:01 -05:00
If no GPU is specified, returns information for all GPUs on the system.
2024-11-11 16:07:11 -06:00
If no topology argument is provided, all topology information will be displayed.
2024-04-15 10:16:39 -07:00
Topology arguments:
2025-05-30 20:08:09 -05:00
-h, --help show this help message and exit
-a, --access Displays link accessibility between GPUs
-w, --weight Displays relative weight between GPUs
-o, --hops Displays the number of hops between GPUs
-t, --link-type Displays the link type between GPUs
-b, --numa-bw Display max and min bandwidth between nodes
-c, --coherent Display cache coherant (or non-coherant) link capability between nodes
-n, --atomics Display 32 and 64-bit atomic io link capability between nodes
-d, --dma Display P2P direct memory access (DMA) link capability between nodes
-z, --bi-dir Display P2P bi-directional link capability between nodes
Device Arguments:
2024-04-15 10:16:39 -07:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-04-15 10:16:39 -07:00
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-set)=
### amd-smi set
Set options for specified devices.
` ``shell-session
2024-06-18 23:09:01 -05:00
~$ amd-smi set --help
2024-04-15 10:16:39 -07:00
usage: amd-smi set [-h] (-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]) [-f %]
[-l LEVEL] [-P SETPROFILE] [-d SCLKMAX] [-C PARTITION] [-M PARTITION]
2024-06-18 23:09:01 -05:00
[-o WATTS] [-p POLICY_ID] [-x POLICY_ID] [-R STATUS]
[--cpu-pwr-limit PWR_LIMIT] [--cpu-xgmi-link-width MIN_WIDTH MAX_WIDTH]
2024-04-15 10:16:39 -07:00
[--cpu-lclk-dpm-level NBIOID MIN_DPM MAX_DPM] [--cpu-pwr-eff-mode MODE]
[--cpu-gmi3-link-width MIN_LW MAX_LW] [--cpu-pcie-link-rate LINK_RATE]
[--cpu-df-pstate-range MAX_PSTATE MIN_PSTATE] [--cpu-enable-apb]
[--cpu-disable-apb DF_PSTATE] [--soc-boost-limit BOOST_LIMIT]
2024-06-18 23:09:01 -05:00
[--core-boost-limit BOOST_LIMIT] [--json | --csv] [--file FILE]
2026-01-06 10:37:07 -06:00
[--loglevel LEVEL] [--cpu-dfcstate-ctrl VALUE] [--cpu-railisofreq-policy VALUE]
2024-04-15 10:16:39 -07:00
2024-11-08 17:31:25 -06:00
If no GPU is specified, will select all GPUs on the system.
2025-03-03 14:53:30 -06:00
A set argument must be provided; Multiple set arguments are accepted.
Requires 'sudo' privileges.
2024-04-15 10:16:39 -07:00
Set Arguments:
2025-05-30 20:08:09 -05:00
-h, --help show this help message and exit
-f, --fan % Set GPU fan speed (0-255 or 0-100%)
-l, --perf-level LEVEL Set one of the following performance levels:
AUTO, LOW, HIGH, MANUAL, STABLE_STD, STABLE_PEAK, STABLE_MIN_MCLK, STABLE_MIN_SCLK, DETERMINISM
-P, --profile PROFILE_LEVEL Set power profile level (#) or choose one of available profiles:
CUSTOM_MASK, VIDEO_MASK, POWER_SAVING_MASK, COMPUTE_MASK, VR_MASK, THREE_D_FULL_SCR_MASK, BOOTUP_DEFAULT
2025-11-17 12:55:14 -06:00
-d, --perf-determinism SCLKMAX Enable performance determinism mode and set GFXCLK softmax limit (in MHz)
2025-05-30 20:08:09 -05:00
-C, --compute-partition TYPE/INDEX Set one of the following the accelerator TYPE or profile INDEX:
N/A.
Use ` sudo amd-smi partition --accelerator` to find acceptable values.
-M, --memory-partition PARTITION Set one of the following the memory partition modes:
NPS1, NPS2, NPS4, NPS8
-o, --power-cap WATTS Set power capacity limit:
min cap: 0 W, max cap: 550 W
-p, --soc-pstate POLICY_ID Set the GPU soc pstate policy using policy id, an integer. Valid id's include:
N/A
-x, --xgmi-plpd POLICY_ID Set the GPU XGMI per-link power down policy using policy id, an integer. Valid id's include:
N/A
-c, --clk-level CLK_TYPE [FREQ_LEVELS ...] Set one or more sclk (aka gfxclk), mclk, fclk, pcie, or socclk frequency levels.
Use ` amd-smi static --clock` to find acceptable levels.
-L, --clk-limit CLK_TYPE LIM_TYPE VALUE Sets the sclk (aka gfxclk) or mclk minimum and maximum frequencies.
ex: amd-smi set -L (sclk | mclk) (min | max) value
-R, --process-isolation STATUS Enable or disable the GPU process isolation on a per partition basis: 0 for disable and 1 for enable.
2025-11-26 08:33:27 -06:00
--ptl-status STATUS Enable or disable the PTL on a GPU processor: 0 for disable and 1 for enable
--ptl-format FRMT1,FRMT2 Set the PTL format on a GPU processor. For example, --ptl-format I8,F32
2024-04-15 10:16:39 -07:00
CPU Arguments:
2025-02-18 15:04:17 -06:00
--cpu-pwr-limit PWR_LIMIT Set power limit for the given socket. Input parameter is power limit value.
--cpu-xgmi-link-width MIN_WIDTH MAX_WIDTH Set max and Min linkwidth. Input parameters are min and max link width values
2025-12-15 13:20:47 -06:00
--cpu-lclk-dpm-level NBIOID MIN_DPM MAX_DPM Sets the max and min dpm level on a given NBIO.
2025-02-18 15:04:17 -06:00
Input parameters are die_index, min dpm, max dpm.
--cpu-pwr-eff-mode MODE Sets the power efficency mode policy. Input parameter is mode.
--cpu-gmi3-link-width MIN_LW MAX_LW Sets max and min gmi3 link width range
--cpu-pcie-link-rate LINK_RATE Sets pcie link rate
--cpu-df-pstate-range MAX_PSTATE MIN_PSTATE Sets max and min df-pstates
--cpu-enable-apb Enables the DF p-state performance boost algorithm
--cpu-disable-apb DF_PSTATE Disables the DF p-state performance boost algorithm. Input parameter is DFPstate (0-3)
--soc-boost-limit BOOST_LIMIT Sets the boost limit for the given socket. Input parameter is socket BOOST_LIMIT value
2026-01-06 10:37:07 -06:00
--cpu-dfcstate-ctrl VALUE Sets the DFCState control for the given socket. Input parameter is VALUE (0-1)
--cpu-railisofreq-policy VALUE Sets the CPU ISO frequency policy. Input parameter is VALUE (0-1)
2024-04-15 10:16:39 -07:00
CPU Core Arguments:
2025-02-18 15:04:17 -06:00
--core-boost-limit BOOST_LIMIT Sets the boost limit for the given core. Input parameter is core BOOST_LIMIT value
2024-04-15 10:16:39 -07:00
2025-05-30 20:08:09 -05:00
Device Arguments:
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
2024-04-15 10:16:39 -07:00
Command Modifiers:
2025-02-18 15:04:17 -06:00
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
2024-04-15 10:16:39 -07:00
` ``
2024-10-02 14:42:34 -04:00
(cmd-reset)=
### amd-smi reset
Reset options for specified devices.
` ``shell-session
2024-04-15 10:16:39 -07:00
~$ amd-smi reset --help
usage: amd-smi reset [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
(-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]) [-G] [-c]
2024-06-18 23:09:01 -05:00
[-f] [-p] [-x] [-d] [-C] [-M] [-o] [-l]
2024-04-15 10:16:39 -07:00
2024-11-08 17:31:25 -06:00
If no GPU is specified, will select all GPUs on the system.
2025-03-03 14:53:30 -06:00
A reset argument must be provided; Multiple reset arguments are accepted.
Requires 'sudo' privileges.
2024-04-15 10:16:39 -07:00
Reset Arguments:
2025-05-30 20:08:09 -05:00
-h, --help show this help message and exit
-G, --gpureset Reset the specified GPU
-c, --clocks Reset clocks and overdrive to default
-f, --fans Reset fans to automatic (driver) control
-p, --profile Reset power profile back to default
-x, --xgmierr Reset XGMI error counts
-d, --perf-determinism Disable performance determinism
-o, --power-cap Reset power capacity limit to max capable
-l, --clean-local-data Clean up local data in LDS/GPRs on a per partition basis
Device Arguments:
2024-04-15 10:16:39 -07:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-04-15 10:16:39 -07:00
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
2024-06-18 23:09:01 -05:00
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-monitor)=
### amd-smi monitor
Monitor metrics for target devices.
` ``shell-session
2024-06-18 23:09:01 -05:00
~$ amd-smi monitor --help
usage: amd-smi monitor [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]]
[-w INTERVAL] [-W TIME] [-i ITERATIONS] [-p] [-t] [-u] [-m] [-n]
2024-06-20 13:05:51 -05:00
[-d] [-e] [-v] [-r] [-q]
2024-06-18 23:09:01 -05:00
Monitor a target device for the specified arguments.
If no arguments are provided, all arguments will be enabled.
2024-11-11 16:07:11 -06:00
Use the watch arguments to run continuously.
2024-06-18 23:09:01 -05:00
Monitor Arguments:
-h, --help show this help message and exit
2025-05-30 20:08:09 -05:00
-p, --power-usage Monitor power usage and power cap in Watts
-t, --temperature Monitor temperature in Celsius
-u, --gfx Monitor graphics utilization (%) and clock (MHz)
-m, --mem Monitor memory utilization (%) and clock (MHz)
-n, --encoder Monitor encoder utilization (%) and clock (MHz)
-d, --decoder Monitor decoder utilization (%) and clock (MHz)
-e, --ecc Monitor ECC single bit, ECC double bit, and PCIe replay error counts
-v, --vram-usage Monitor memory usage in MB
-r, --pcie Monitor PCIe bandwidth in Mb/s
-q, --process Enable Process information table below monitor output
-V, --violation Monitor power and thermal violation status (%);
Only available for MI300 or newer ASICs
Watch Arguments:
-w, --watch INTERVAL Reprint the command in a loop of INTERVAL seconds
-W, --watch_time TIME The total duration of TIME to watch the command
-i, --iterations ITERATIONS The total number of ITERATIONS to repeat the command
Device Arguments:
2024-06-18 23:09:01 -05:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-06-18 23:09:01 -05:00
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2024-10-02 14:42:34 -04:00
(cmd-xgmi)=
### amd-smi xgmi
Displays XGMI information of specified devices.
` ``shell-session
2024-06-18 23:09:01 -05:00
~$ amd-smi xgmi --help
usage: amd-smi xgmi [-h] [--json | --csv] [--file FILE] [--loglevel LEVEL]
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]] [-m]
If no GPU is specified, returns information for all GPUs on the system.
2024-11-11 16:07:11 -06:00
If no xgmi argument is provided, all xgmi information will be displayed.
2024-06-18 23:09:01 -05:00
XGMI arguments:
2025-05-30 20:08:09 -05:00
-h, --help show this help message and exit
-m, --metric Metric XGMI information
-l, --link-status XGMI Link Status information
Device Arguments:
2024-06-18 23:09:01 -05:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
2025-02-18 15:04:17 -06:00
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
2024-06-18 23:09:01 -05:00
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
2024-04-15 10:16:39 -07:00
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2025-03-13 05:59:34 -04:00
(cmd-partition)=
### amd-smi partition
Displays partition information of the devices.
` ``shell-session
~$ amd-smi partition --help
usage: amd-smi partition [-h] [-g GPU [GPU ...]] [-c] [-m] [-a] [--json | --csv]
[--file FILE] [--loglevel LEVEL]
If no GPU is specified, returns information for all GPUs on the system.
If no partition argument is provided, all partition information will be displayed.
2025-05-08 12:19:04 -05:00
Partition arguments:
2025-03-13 05:59:34 -04:00
-h, --help show this help message and exit
2025-05-30 20:08:09 -05:00
-c, --current display the current partition information
-m, --memory display the current memory partition mode and capabilities
-a, --accelerator display accelerator partition information
Device Arguments:
2025-03-13 05:59:34 -04:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
2025-05-08 12:19:04 -05:00
` ``
### amd-smi ras
Displays RAS information of specified devices.
` ``shell-session
~$ amd-smi ras --help
usage: amd-smi ras [-h] --cper [--severity SEVERITY [SEVERITY ...]] [--folder FOLDER]
2025-05-29 11:59:55 -05:00
[--file-limit FILE_LIMIT] [--follow]
2025-05-08 12:19:04 -05:00
[-g GPU [GPU ...] | -U CPU [CPU ...] | -O CORE [CORE ...]]
[--json | --csv] [--file FILE] [--loglevel LEVEL]
2025-05-30 20:08:09 -05:00
Retrieve and decode RAS (CPER) entries from the kernel driver.
2025-05-08 12:19:04 -05:00
Supports filtering by severity, exporting to different formats, and continuous monitoring.
This command accepts options only; no positional arguments are required.
RAS arguments:
-h, --help show this help message and exit
--cper Trigger CPER data retrieval
2025-08-19 18:55:33 -05:00
--afid Generate an AFID (AMD Field ID) given a CPER record file.
2025-05-08 12:19:04 -05:00
--severity SEVERITY [SEVERITY ...] Set the SEVERITY filters from the following:
nonfatal-uncorrected, fatal, nonfatal-corrected, all
--folder FOLDER Folder to dump CPER report files
2025-05-29 11:59:55 -05:00
--file-limit FILE_LIMIT Maximum number of entries per output file
2025-08-19 18:55:33 -05:00
--cper-file CPER_FILE Full path of the CPER record file to generate the AFID
2025-05-08 12:19:04 -05:00
--follow Continuously monitor for new entries
2025-05-30 20:08:09 -05:00
Device Arguments:
2025-05-08 12:19:04 -05:00
-g, --gpu GPU [GPU ...] Select a GPU ID, BDF, or UUID from the possible choices:
ID: 0 | BDF: 0000:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 1 | BDF: 0001:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 2 | BDF: 0002:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
ID: 3 | BDF: 0003:01:00.0 | UUID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
all | Selects all devices
-U, --cpu CPU [CPU ...] Select a CPU ID from the possible choices:
ID: 0
ID: 1
ID: 2
ID: 3
all | Selects all devices
-O, --core CORE [CORE ...] Select a Core ID from the possible choices:
ID: 0 - 95
all | Selects all devices
Command Modifiers:
--json Displays output in JSON format (human readable by default).
--csv Displays output in CSV format (human readable by default).
--file FILE Saves output into a file on the provided path (stdout by default).
--loglevel LEVEL Set the logging level from the possible choices:
2025-03-13 05:59:34 -04:00
DEBUG, INFO, WARNING, ERROR, CRITICAL
` ``
2025-05-14 12:43:33 -04:00
## Interpreting the output
When you run an ` amd-smi` command, the tool presents detailed information
across various categories, each containing specific fields and their current
values.
(cli-output-na)=
### About N/A values
` N/A` typically indicates that the data for a specific field is either unavailable or irrelevant in the current context for your device and its software environment. The exact reason may vary depending on the field and the GPU. In general, you can interpret ` N/A` to mean one of the following:
**Not Applicable**: The feature or parameter does not apply to your specific AMD hardware or its current configuration. Examples include display-related clocks on a headless compute card and partition details when the GPU is not partitioned.
**Not Available**: The information cannot be retrieved by the ` amd-smi` tool at this time. This could be due to one of the following reasons:
- The hardware component does not report the specific metric.
- The currently installed ` amdgpu` driver version does not support querying this particular piece of information through ` amd-smi-lib`.
2024-10-02 14:42:34 -04:00
(cli-ex-static)=
2024-04-15 10:16:39 -07:00
### Example output from amd-smi static
2025-05-14 12:43:33 -04:00
The following block is example output from the ` amd-smi static` command without additional modifiers.
2024-04-15 10:16:39 -07:00
` ``bash
~$ amd-smi static
CPU: 0
2024-06-18 23:09:01 -05:00
SMU:
FW_VERSION: 85.90.0
INTERFACE_VERSION:
PROTO VERSION: 6
2024-11-11 16:07:11 -06:00
...
2024-04-15 10:16:39 -07:00
GPU: 0
ASIC:
2025-02-18 15:04:17 -06:00
MARKET_NAME: AMD Instinct MI300A
2024-04-15 10:16:39 -07:00
VENDOR_ID: 0x1002
VENDOR_NAME: Advanced Micro Devices Inc. [AMD/ATI]
SUBVENDOR_ID: 0x1002
DEVICE_ID: 0x74a0
2024-11-11 16:07:11 -06:00
SUBSYSTEM_ID: 0x74a0
2024-06-18 23:09:01 -05:00
REV_ID: 0x00
2025-05-30 20:08:09 -05:00
ASIC_SERIAL: 0xXXXXXXXXXXXXXXXX
2024-04-15 10:16:39 -07:00
OAM_ID: 0
2024-11-11 16:07:11 -06:00
NUM_COMPUTE_UNITS: 228
TARGET_GRAPHICS_VERSION: gfx942
2024-04-15 10:16:39 -07:00
BUS:
BDF: 0000:01:00.0
MAX_PCIE_WIDTH: 16
MAX_PCIE_SPEED: 32 GT/s
PCIE_INTERFACE_VERSION: Gen 5
SLOT_TYPE: PCIE
2025-09-16 19:51:13 -05:00
IFWI:
2024-04-15 10:16:39 -07:00
NAME: N/A
BUILD_DATE: N/A
PART_NUMBER: N/A
2024-06-18 23:09:01 -05:00
VERSION: N/A
2024-04-15 10:16:39 -07:00
LIMIT:
MAX_POWER: 550 W
2024-06-18 23:09:01 -05:00
MIN_POWER: 0 W
2024-04-15 10:16:39 -07:00
SOCKET_POWER: 550 W
SLOWDOWN_EDGE_TEMPERATURE: N/A
SLOWDOWN_HOTSPOT_TEMPERATURE: 100 °C
2024-06-18 23:09:01 -05:00
SLOWDOWN_VRAM_TEMPERATURE: 105 °C
2024-04-15 10:16:39 -07:00
SHUTDOWN_EDGE_TEMPERATURE: N/A
SHUTDOWN_HOTSPOT_TEMPERATURE: 110 °C
2024-06-18 23:09:01 -05:00
SHUTDOWN_VRAM_TEMPERATURE: 115 °C
2024-04-15 10:16:39 -07:00
DRIVER:
NAME: amdgpu
2025-05-30 20:08:09 -05:00
VERSION: 6.14.4
2024-04-15 10:16:39 -07:00
BOARD:
MODEL_NUMBER: N/A
PRODUCT_SERIAL: N/A
FRU_ID: N/A
2024-06-18 23:09:01 -05:00
PRODUCT_NAME: Aqua Vanjaram [Instinct MI300A]
MANUFACTURER_NAME: Advanced Micro Devices, Inc. [AMD/ATI]
2024-04-15 10:16:39 -07:00
RAS:
2025-05-30 20:08:09 -05:00
EEPROM_VERSION: 0x30000
BAD_PAGE_THRESHOLD: N/A
2024-04-15 10:16:39 -07:00
PARITY_SCHEMA: DISABLED
SINGLE_BIT_SCHEMA: DISABLED
DOUBLE_BIT_SCHEMA: DISABLED
POISON_SCHEMA: ENABLED
ECC_BLOCK_STATE:
UMC: DISABLED
SDMA: ENABLED
GFX: ENABLED
MMHUB: ENABLED
ATHUB: DISABLED
PCIE_BIF: DISABLED
HDP: DISABLED
XGMI_WAFL: DISABLED
DF: DISABLED
SMN: DISABLED
SEM: DISABLED
MP0: DISABLED
MP1: DISABLED
FUSE: DISABLED
2024-06-18 23:09:01 -05:00
MCA: DISABLED
VCN: DISABLED
JPEG: DISABLED
IH: DISABLED
MPIO: DISABLED
2024-04-15 10:16:39 -07:00
PARTITION:
2025-05-30 20:08:09 -05:00
ACCELERATOR_PARTITION: SPX
2024-04-15 10:16:39 -07:00
MEMORY_PARTITION: NPS1
2024-11-11 16:07:11 -06:00
PARTITION_ID: 0
2025-02-18 15:04:17 -06:00
SOC_PSTATE: N/A
XGMI_PLPD: N/A
PROCESS_ISOLATION: Disabled
2024-04-15 10:16:39 -07:00
NUMA:
NODE: 0
AFFINITY: 0
2025-05-30 20:08:09 -05:00
CPU AFFINITY:
0xffffff
0xffffff00000000
0x0
SOCKET AFFINITY:
0
2024-04-15 10:16:39 -07:00
VRAM:
TYPE: HBM
2025-02-18 15:04:17 -06:00
VENDOR: UNKNOWN
2025-05-30 20:08:09 -05:00
SIZE: 96432 MB
2024-11-11 16:07:11 -06:00
BIT_WIDTH: 8192
2025-05-30 20:08:09 -05:00
MAX_BANDWIDTH: 5325 GB/s
2024-06-18 23:09:01 -05:00
CACHE_INFO:
CACHE_0:
CACHE_PROPERTIES: DATA_CACHE, SIMD_CACHE
CACHE_SIZE: 32 KB
CACHE_LEVEL: 1
2025-05-30 20:08:09 -05:00
MAX_NUM_CU_SHARED: 1
NUM_CACHE_INSTANCE: 228
2024-06-18 23:09:01 -05:00
CACHE_1:
CACHE_PROPERTIES: INST_CACHE, SIMD_CACHE
CACHE_SIZE: 64 KB
CACHE_LEVEL: 1
MAX_NUM_CU_SHARED: 2
2025-05-30 20:08:09 -05:00
NUM_CACHE_INSTANCE: 108
2024-06-18 23:09:01 -05:00
CACHE_2:
2025-05-30 20:08:09 -05:00
CACHE_PROPERTIES: INST_CACHE, SIMD_CACHE
CACHE_SIZE: 64 KB
CACHE_LEVEL: 1
MAX_NUM_CU_SHARED: 1
NUM_CACHE_INSTANCE: 12
CACHE_3:
CACHE_PROPERTIES: DATA_CACHE, SIMD_CACHE
CACHE_SIZE: 16 KB
CACHE_LEVEL: 1
MAX_NUM_CU_SHARED: 2
NUM_CACHE_INSTANCE: 108
CACHE_4:
CACHE_PROPERTIES: DATA_CACHE, SIMD_CACHE
CACHE_SIZE: 16 KB
CACHE_LEVEL: 1
MAX_NUM_CU_SHARED: 1
NUM_CACHE_INSTANCE: 12
CACHE_5:
2024-06-18 23:09:01 -05:00
CACHE_PROPERTIES: DATA_CACHE, SIMD_CACHE
CACHE_SIZE: 4096 KB
CACHE_LEVEL: 2
MAX_NUM_CU_SHARED: 228
NUM_CACHE_INSTANCE: 1
2025-05-30 20:08:09 -05:00
CACHE_6:
2024-06-18 23:09:01 -05:00
CACHE_PROPERTIES: DATA_CACHE, SIMD_CACHE
CACHE_SIZE: 262144 KB
CACHE_LEVEL: 3
MAX_NUM_CU_SHARED: 228
NUM_CACHE_INSTANCE: 1
2024-12-04 00:10:32 -06:00
CLOCK:
SYS:
2025-02-18 15:04:17 -06:00
CURRENT LEVEL: 1
2024-12-04 00:10:32 -06:00
FREQUENCY_LEVELS:
2025-05-30 20:08:09 -05:00
LEVEL 0: 500 MHz
LEVEL 1: 207 MHz
LEVEL 2: 2100 MHz
2025-02-18 15:04:17 -06:00
MEM:
2025-05-30 20:08:09 -05:00
CURRENT LEVEL: 3
2025-02-18 15:04:17 -06:00
FREQUENCY_LEVELS:
LEVEL 0: 900 MHz
LEVEL 1: 1100 MHz
LEVEL 2: 1200 MHz
LEVEL 3: 1300 MHz
2024-12-04 00:10:32 -06:00
DF:
2025-05-30 20:08:09 -05:00
CURRENT LEVEL: 3
2024-12-04 00:10:32 -06:00
FREQUENCY_LEVELS:
2025-02-18 15:04:17 -06:00
LEVEL 0: 1200 MHz
LEVEL 1: 1600 MHz
LEVEL 2: 1900 MHz
LEVEL 3: 2000 MHz
2024-12-04 00:10:32 -06:00
SOC:
CURRENT LEVEL: 0
FREQUENCY_LEVELS:
2025-05-30 20:08:09 -05:00
LEVEL 0: 43 MHz
2025-02-18 15:04:17 -06:00
LEVEL 1: 800 MHz
LEVEL 2: 1000 MHz
LEVEL 3: 1143 MHz
DCEF: N/A
VCLK0:
CURRENT LEVEL: 0
FREQUENCY_LEVELS:
2025-05-30 20:08:09 -05:00
LEVEL 0: 54 MHz
2025-02-18 15:04:17 -06:00
VCLK1:
CURRENT LEVEL: 0
FREQUENCY_LEVELS:
2025-05-30 20:08:09 -05:00
LEVEL 0: 54 MHz
2025-02-18 15:04:17 -06:00
DCLK0:
CURRENT LEVEL: 0
FREQUENCY_LEVELS:
2025-05-30 20:08:09 -05:00
LEVEL 0: 45 MHz
2025-02-18 15:04:17 -06:00
DCLK1:
2024-12-04 00:10:32 -06:00
CURRENT LEVEL: 0
FREQUENCY_LEVELS:
2025-05-30 20:08:09 -05:00
LEVEL 0: 45 MHz
2024-11-11 16:07:11 -06:00
...
2024-04-15 10:16:39 -07:00
` ``
2025-09-09 11:27:15 -04:00
### Listing CPER entries using amd-smi
2025-09-16 18:53:07 -05:00
This example code shows how to list CPER entries for all GPUs into files
` ``bash
~$ sudo amd-smi ras --cper --severity all --folder /tmp/cper_dump/
timestamp gpu_id severity file_name list of afids
2000/06/27 10:45:13 0 FATAL fatal-1.cper 30
2000/06/27 10:45:13 1 FATAL fatal-2.cper 30
2000/06/27 10:45:13 2 FATAL fatal-3.cper 30
2000/06/27 10:45:13 3 FATAL fatal-4.cper 30
2000/06/27 10:45:13 4 FATAL fatal-5.cper 30
2000/06/27 10:45:13 5 FATAL fatal-6.cper 30
2000/06/27 10:45:13 6 FATAL fatal-7.cper 30
2000/06/27 10:45:13 7 FATAL fatal-8.cper 30
` ``
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
This example code shows how to list CPER entries for a given GPU into files
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
` ``bash
~$ sudo amd-smi ras --cper --severity all --folder /tmp/cper_dump/ --gpu 1
timestamp gpu_id severity file_name list of afids
2000/06/27 10:45:13 1 FATAL fatal-1.cper 30
2025-09-09 11:27:15 -04:00
` ``
2025-09-16 18:53:07 -05:00
This example code shows how to list CPER entries and their JSON data for a given GPU into files
` ``bash
~$ sudo amd-smi ras --cper --severity all --folder /tmp/cper_dump/ --gpu 1 --json
timestamp gpu_id severity file_name list of afids
2000/06/27 10:45:13 1 FATAL fatal-1.cper 30
~$ ls -alh /tmp/cper_dump/
total 12K
drwxr-xr-x 2 root root 46 Sep 16 21:12 .
drwxrwxrwt 1 root root 4.0K Sep 16 18:03 ..
-rw-r--r-- 1 root root 376 Sep 16 21:12 fatal-1.cper
-rw-r--r-- 1 root root 347 Sep 16 21:12 fatal-1.json
~$ cat /tmp/cper_dump/fatal-1.json
{
"error_severity": "fatal",
"notify_type": "MCE",
"timestamp": "2000/06/27 10:45:13",
"signature": "CPER",
"revision": 256,
"signature_end": "0xffffffff",
"sec_cnt": 1,
"record_length": 376,
"platform_id": "111102-G40307-0C",
"creator_id": "136c692517001839",
"record_id": "f0000031",
"flags": 0,
"persistence_info": 0
}
2025-09-09 11:27:15 -04:00
` ``
2025-09-16 18:53:07 -05:00
This example code shows how to continuously list CPER entries without exiting
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
` ``bash
~$ sudo amd-smi ras --cper --follow --severity all --folder /tmp/cper_dump
Press CTRL + C to stop.
timestamp gpu_id severity file_name list of afids
2000/06/27 10:45:13 0 FATAL fatal-1.cper 30
2000/06/27 10:45:13 1 FATAL fatal-2.cper 30
2000/06/27 10:45:13 2 FATAL fatal-3.cper 30
2000/06/27 10:45:13 3 FATAL fatal-4.cper 30
2000/06/27 10:45:13 4 FATAL fatal-5.cper 30
2000/06/27 10:45:13 5 FATAL fatal-6.cper 30
2000/06/27 10:45:13 6 FATAL fatal-7.cper 30
2000/06/27 10:45:13 7 FATAL fatal-8.cper 30
...
` ``
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
This example code shows how to list CPER entries with a limited number of entries
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
` ``bash
~$ sudo amd-smi ras --cper --severity all --folder /tmp/cper_dump --file-limit 5
timestamp gpu_id severity file_name list of afids
2000/06/27 10:45:13 0 FATAL fatal-1.cper 30
2000/06/27 10:45:13 1 FATAL fatal-2.cper 30
2000/06/27 10:45:13 2 FATAL fatal-3.cper 30
2000/06/27 10:45:13 3 FATAL fatal-4.cper 30
2000/06/27 10:45:13 4 FATAL fatal-5.cper 30
` ``
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
This example code shows how to list a specific severity of CPER entries only
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
` ``bash
~$ sudo amd-smi ras --cper --severity fatal --folder /tmp/cper_dump/
timestamp gpu_id severity file_name list of afids
2000/06/27 10:45:13 0 FATAL fatal-1.cper 30
2025-09-09 11:27:15 -04:00
` ``
2025-09-16 18:53:07 -05:00
This example code shows how to dump AFID errors in a CPER file
2025-09-09 11:27:15 -04:00
2025-09-16 18:53:07 -05:00
` ``bash
~$ sudo amd-smi ras --afid --cper-file /tmp/cper_dump/fatal-1.cper
2025-09-09 11:27:15 -04:00
` ``
2025-09-19 11:07:38 -04:00
Refer to
[amd_smi_cper_example.py ](https://github.com/ROCm/amdsmi/blob/amd-mainline/example/amd_smi_cper_example.py )
and
[amd_smi_afid_example.py ](https://github.com/ROCm/amdsmi/blob/amd-mainline/example/amd_smi_afid_example.py )
for API examples.
