AMD GPU metrics
Shows hardware metrics for AMD GPUs
| Metric name | Type | Description |
|---|---|---|
| GPU_NODES_TOTAL | Number of GPU nodes on the machine | |
| GPU_PACKAGE_POWER | Current socket power in Watts; not available on guest VM | |
| GPU_AVERAGE_PACKAGE_POWER | Average socket power in Watts; not available on guest VM | |
| GPU_EDGE_TEMPERATURE | Edge temperature value in Celsius | |
| GPU_JUNCTION_TEMPERATURE | Hotspot (aka junction) temperature value in Celsius | |
| GPU_MEMORY_TEMPERATURE | Memory temperature value in Celsius | |
| GPU_HBM_TEMPERATURE | List of hbm temperatures in Celsius | |
| GPU_GFX_ACTIVITY | Graphics engine usage percentage (0 - 100) | |
| GPU_UMC_ACTIVITY | Memory engine usage percentage (0 - 100) | |
| GPU_MMA_ACTIVITY | Average multimedia engine usages in percentage (0 - 100) | |
| GPU_VCN_ACTIVITY | List of VCN encode/decode engine utilization per AID | |
| GPU_JPEG_ACTIVITY | List of JPEG engine activity in percentage (0 - 100) | |
| GPU_VOLTAGE | SoC voltage in mV | |
| GPU_GFX_VOLTAGE | gfx voltage in mV | |
| GPU_MEMORY_VOLTAGE | Mem voltage in mV | |
| PCIE_SPEED | Current pcie speed capable in GT/s | |
| PCIE_MAX_SPEED | Maximum capable pcie speed in GT/s | |
| PCIE_BANDWIDTH | Current instantaneous bandwidth usage in Mb/s | |
| GPU_ENERGY_CONSUMED | Energy consumed by GPU in Micro Jules (uJ) | |
| PCIE_REPLAY_COUNT | Total number of PCIe replays (NAKs) | |
| PCIE_RECOVERY_COUNT | Total number of PCIe replays (NAKs) | |
| PCIE_REPLAY_ROLLOVER_COUNT | PCIe Replay accumulated count | |
| PCIE_NACK_SENT_COUNT | PCIe NAK sent accumulated count | |
| PCIE_NAC_RECEIVED_COUNT | PCIe NAK received accumulated count | |
| GPU_CLOCK | Clock measure of the GPU in Mhz | |
| GPU_POWER_USAGE | GPU power usage in Watts | |
| GPU_TOTAL_VRAM | Total VRAM available in MB | |
| GPU_ECC_CORRECT_TOTAL | Total Correctable ECC error count | |
| GPU_ECC_UNCORRECT_TOTAL | Total Uncorrectable ECC error count | |
| GPU_ECC_CORRECT_SDMA | Correctable ECC error in SDMA | |
| GPU_ECC_UNCORRECT_SDMA | Uncorrectable ECC error in SDMA | |
| GPU_ECC_CORRECT_GFX | Correctable ECC error in GFX | |
| GPU_ECC_UNCORRECT_GFX | Uncorrectable ECC error in GFX | |
| GPU_ECC_CORRECT_MMHUB | Correctable ECC error in MMHUB | |
| GPU_ECC_UNCORRECT_MMHUB | Uncorrectable ECC error in MMHUB | |
| GPU_ECC_CORRECT_ATHUB | Correctable ECC error in ATHUB | |
| GPU_ECC_UNCORRECT_ATHUB | Uncorrectable ECC error in ATHUB | |
| GPU_ECC_CORRECT_BIF | Correctable ECC error in BIF | |
| GPU_ECC_UNCORRECT_BIF | Uncorrectable ECC error in BIF | |
| GPU_ECC_CORRECT_HDP | Correctable ECC error in HDP | |
| GPU_ECC_UNCORRECT_HDP | Uncorrectable ECC error in HDP | |
| GPU_ECC_CORRECT_XGMI_WAFL | Correctable ECC error in XGMI WAFL | |
| GPU_ECC_UNCORRECT_XGMI_WAFL | Uncorrectable ECC error in XGMI WAFL | |
| GPU_ECC_CORRECT_DF | Correctable ECC error in DF | |
| GPU_ECC_UNCORRECT_DF | Uncorrectable ECC error in DF | |
| GPU_ECC_CORRECT_SMN | Correctable ECC error in SMN | |
| GPU_ECC_UNCORRECT_SMN | Uncorrectable ECC error in SMN | |
| GPU_ECC_CORRECT_SEM | Correctable ECC error in SEM | |
| GPU_ECC_UNCORRECT_SEM | Uncorrectable ECC error in SEM | |
| GPU_ECC_CORRECT_MP0 | Correctable ECC error in MP0 | |
| GPU_ECC_UNCORRECT_MP0 | Uncorrectable ECC error in MP0 | |
| GPU_ECC_CORRECT_MP1 | Correctable ECC error in MP1 | |
| GPU_ECC_UNCORRECT_MP1 | Uncorrectable ECC error in MP1 | |
| GPU_ECC_CORRECT_FUSE | Correctable ECC error in FUSE | |
| GPU_ECC_UNCORRECT_FUSE | Uncorrectable ECC error in FUSE | |
| GPU_ECC_CORRECT_UMC | Correctable ECC error in UMC | |
| GPU_ECC_UNCORRECT_UMC | Uncorrectable ECC error in UMC |