GitHub Hot Board

GitHub Hot Board · Arsip repositori

intel/metrics-discovery

Belum ada deskripsi untuk bahasa ini — muncul otomatis.

C++

Data per October 28intel

Peringkat terbaik

#214

Permata yang Terlupakan

100

Total bintang

39

Total fork

5

Hari di papan (30 hari)

Lisensi: MITPush terakhir: October 27Issue terbuka (termasuk PR): 1Dikumpulkan sejak 2026-10-25; tidak diketahui = belum terkumpul

Dokumentasi proyek

Intel(R) Metrics Discovery Application Programming Interface

Introduction

This software is a user mode library that provides access to GPU performance data.

The library has OS abstraction layer that allows for interoperability in different environments.

License

Intel(R) Metrics Discovery Application Programming Interface is distributed under the MIT License.

You may obtain a copy of the License at: https://opensource.org/licenses/MIT

Supported Platforms

  • Intel(R) Processors with Xe3P graphics devices (formerly Crescent Island, Nova Lake-P)
  • Intel(R) Processors with Xe3 graphics devices (formerly Panther Lake, Wildcat Lake, Nova Lake)
  • Intel(R) Processors with Xe2 graphics devices (formerly Lunar Lake)
  • Intel(R) Processors with Arc graphics devices (formerly Alchemist, Battlemage)
  • Intel(R) Processors with Xe graphics devices (formerly Tiger Lake, Rocket Lake, DG1, Alder Lake, Ponte Vecchio, Meteor Lake, Arrow Lake)

Since version 1.14.181, support for Gen7.5, Gen8, Gen9 and Gen11 platforms has been discontinued. For these platforms, use version 1.14.180 or earlier.

Supported Operating Systems

Intel(R) Metrics Discovery Application Programming Interface is supported on Linux family operating systems with minimum kernel version 5.7. Event based measurements require minimum Mesa version 18.2.

Detailed minimum version requirements

Platform Code Name Minimum Linux Kernel Version Minimum Mesa Version
Xe3P-LPG Nova Lake-P 7.1 26.1
Xe3P-XPC Crescent Island 6.19 ----
Xe3-LPG Nova Lake 6.19 26.0
Xe3-LPG Wildcat Lake 6.18 25.3
Xe3-LPG Panther Lake 6.15 25.1
Xe2-HPG Battlemage 6.11 24.2
Xe2-LPG Lunar Lake 6.11 24.2
Xe-LPG Arrow Lake-H 6.9 24.1
Xe-LPG Arrow Lake-S 6.7 24.1
Xe-LPG Arrow Lake-U 6.9 24.1
Xe-LPG Meteor Lake 6.7 23.1
Xe-HPG Alchemist 6.2 22.2
Xe-HPC Ponte Vecchio 5.20 18.2
Xe-LP Alder Lake-S 5.16 22.0
Xe-LP Alder Lake-N 5.18 22.0
Xe-LP Alder Lake-P 5.17 22.0
Xe-LP DG1 5.16 22.1.5
Xe-LP Rocket Lake 5.13 20.2
Xe-LP Tiger Lake 5.7 21.0

Build and Install

1. Download sources.

2. Install libdrm-dev.

sudo apt install libdrm-dev

or

sudo dnf install libdrm-devel

3. Run CMake generation:

mkdir build && cd build
cmake ..

4. Build:

make -j$(nproc)

5. Built library will be here (for 64-bit Linux):

(project_root)/dump/linux64/release/metrics_discovery/libigdmd.so

6. Install:

If using native package manager (yum, dnf, etc..) skip to step 7

sudo make install

7. Packaging:

make package

After the process completes, there should be a set of files such as

intel-metrics-discovery-[VERSION].x86_64.rpm
intel-metrics-discovery-devel-[VERSION].x86_64.rpm
intel-metrics-discovery-pkgconfig-[VERSION].x86_64.rpm

Note: To clear CMake params remove CMakeCache.txt, then regenerate.

Support

Please file a GitHub issue to report an issue or ask questions.

How to Contribute

Contributions to the Metrics-Discovery are welcomed and encouraged. Please see CONTRIBUTING for details how to contribute to the project.

(*) Other names and brands my be claimed as property of others.

Peringkat saat ini

  • 214

    Permata yang Terlupakan

    100 bintang · 39 fork · Total Star (menaik)

    Terlupakan →

Grafik 30 hari

Jumlah bintang kumulatif

Pergerakan peringkat

Kenaikan harian

Pasang lencana di README kamu · intel/metrics-discovery[![Pasang lencana di README kamu](https://git.vaiox.xyz/api/badge/intel/metrics-discovery.svg)](https://git.vaiox.xyz/id/repo/intel/metrics-discovery)

Badge · 任何仓库都可以挂 —— 未上榜的仓库显示灰章,上完榜会自动点亮。

<a href="https://git.vaiox.xyz/id/repo/intel/metrics-discovery"><img src="https://git.vaiox.xyz/api/badge/intel/metrics-discovery.svg" alt="Pasang lencana di README kamu" height="20"></a>

← Kembali ke papan

GitHub Hot BoardMesin WaktuTemukan联系我们 · Contact