Bugfix

• Snapshot mode (-s) outputs valid JSON. Thanks @Steve-Tech

Additions

• Loop Snapshot mode (-l) prints snapshots in a loop fashion. Thanks @Steve-Tech
• Extra snapshot information - @Syllo
  • Processes
  • Encode/decode

• Fix a regression that would miss-report available device memory of NVIDIA GPUs
• Integrate effective/efficiency usage within the GPU utilization metric

New in this release

Target support

• Rockchip NPU - @feilongfl
• MetaX GPUs - @zhenyu-xu-metax
• Enflame GCU - @QShen3

Improvements

• More memory reporting in one-shot mode - @mintyleaf
• CTRL + L will reset the ncurses interface - @cipri-tom + @claude
• GPU and MEM have separate clock fields - @KaeLL
• Effective load metric, i.e., the percentage utilization weighted...

New this release

• Intel XE driver support thanks to @Steve-Tech
  • Supports more metrics when CAP_PERFMON is given to nvtop. See the intel section of the README
• Broadcom V3D support (Raspberry Pi) thanks to @xiaohuirong
• Google TPU support thanks to @rdyro
• Interface hide process list with the -P option thanks to @rin...

Added support for the following hardware accelerators/GPUs:

• More Adreno GPUs thanks to the panfrost linux driver
• Mac GPUs (tested on M1 and M2 by the community)
• Huawei Ascend accelerators

Bugfixes and improvements:

• Don't call handlers for GPUs that are marked as not to be monitored
• Fix crash when the configuration file path could not be discovered through user environme...