Skip to content

Latest commit

 

History

History
280 lines (157 loc) · 10.2 KB

File metadata and controls

280 lines (157 loc) · 10.2 KB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.


Added

Changed

Fixed

Removed


1.6.2 - 2026-01-28

Added

  • Add nvidia-ml-py 13.580.126 and 13.590.44 and 13.590.48 to support list by @XuehaiPan.

Fixed

  • Fix TUI column alignment for device with power limit >= 1000W by @XuehaiPan.
  • Fix grammar and typos in documentation by @XuehaiPan.

1.6.1 - 2025-12-08

Added


1.6.0 - 2025-11-11

Added

  • Add nvidia-ml-py 13.580.82 to support list for NVIDIA Spark/Thor by @johnnynunez in #186.
  • Add bar charts for memory bandwidth and power usage in the main screen by @XuehaiPan in #190.
  • Add workflow to build and publish Docker images to GitHub Container Registry by @XuehaiPan in #190. Issued by @ntheanh201.

1.5.3 - 2025-08-16

Added

Changed

  • Draw network and disk I/O graphs with centered symmetric zero in Grafana dashboard by @XuehaiPan.

Fixed

  • Fix device.pcie_tx_throughput() returns PCIe RX throughput due to a typo in argument by @kyet in #176.

Removed

  • Remove per-version install extras for nvidia-ml-py and prefer nvitop[cudaXX] instead by @XuehaiPan in #179.

1.5.2 - 2025-07-25

Changed

  • Minor tweak display for device name in TUI by @XuehaiPan.

Fixed

  • Fix snapshot cache for GPU processes with shared host process by @XuehaiPan in #172.

1.5.1 - 2025-05-26

Added

Fixed

  • Fix one-time output rendering on exit for TUI by @XuehaiPan.

1.5.0 - 2025-04-25

Added

Changed

  • Deprecate nvitop.callbacks as officially unmaintained by @XuehaiPan in #157.

Fixed

  • Ignore errors when collecting host metrics for host panel by @XuehaiPan in #163.

Removed


1.4.2 - 2025-01-27

Removed


1.4.1 - 2025-01-13

Fixed

  • Fix passing invalid device handle (e.g., GPU is lost) to NVML functions by @XuehaiPan in #146.
  • Fix CUDA device selection tool nvisel by @XuehaiPan.

1.4.0 - 2024-12-29

Added

Changed

Fixed


1.3.2 - 2023-10-17

Added

  • Add separate implementation for GpuStatsLogger callback for lightning by @XuehaiPan in #114.
  • Remove metrics if process is gone in nvitop-exporter by @XuehaiPan in #107.

1.3.1 - 2023-10-05

Added

Fixed

  • Fix libcuda.cuDeviceGetUuid() when the UUID contains 0x00 by @XuehaiPan in #100.

1.3.0 - 2023-08-27

Added

Changed

  • Use recent timestamp for GPU process utilization query for more accurate per-process GPU usage by @XuehaiPan in #85. We extend our heartfelt gratitude to @2581543189 for their invaluable assistance. Their timely comments and comprehensive feedback have greatly contributed to the improvement of this project.

Fixed

  • Fix upstream changes for process info v3 APIs on 535.104.05 driver by @XuehaiPan in #94.
  • Fix removal for process info v3 APIs on the upstream 535.98 driver by @XuehaiPan in #89.

1.2.0 - 2023-07-24

Added

  • Include last snapshot metrics in the log results for ResourceMetricCollector by @XuehaiPan in #80.
  • Add mypy integration and update type annotations by @XuehaiPan in #73.

Fixed

  • Fix process info support for NVIDIA R535 driver (CUDA 12.2+) by @XuehaiPan in #79.
  • Fix inappropriate exception catching in function libcuda.cuDeviceGetUuid by @XuehaiPan.

1.1.2 - 2023-04-11

Fixed

  • Further isolate the CUDA_VISIBLE_DEVICES parser in a subprocess by @XuehaiPan in #70.

1.1.1 - 2023-04-07

Fixed


1.1.0 - 2023-04-07

Added

  • Support float number as snapshot interval that >= 0.25s by @XuehaiPan in #67.
  • Show more host metrics (e.g., used virtual memory, uptime) in CLI by @XuehaiPan in #59.

Changed

Fixed

  • Respect FORCE_COLOR and NO_COLOR environment variables by @XuehaiPan.

Removed


1.0.0 - 2023-02-01

Added

  • The first stable release of nvitop by @XuehaiPan.