Sun, 15 Nov 2026 13:07:33 UTC Log in
Measure the peak achievable performance of GPU compute devices 1.09 MB built 2026-10-13 05:27:51 GMT

Description

A synthetic micro-benchmark that measures the peak achievable performance of GPU compute devices. It exercises tight vector / MAD / MMA loops and vendor-SDK GEMM libraries (cuBLASLt on NVIDIA, MPS on Apple) to expose what the hardware is capable of — from raw ALU peaks to near-vendor-advertised matrix throughput. clpeak began as an OpenCL-only tool. It now ships four interchangeable backends — OpenCL, Vulkan, CUDA, and Metal — running back-to-back on the same hardware, so cross-stack differences (driver lowering, instruction scheduling, extension exposure) become visible alongside the raw peak numbers.

Dependencies

12 total
Requires11
  • cmake
  • gcc
  • gcc-c++
  • glslc
  • opencl-headers
  • pkgconfig(OpenCL)
  • pkgconfig(gl)
  • pkgconfig(vulkan)
  • rpmlib(CompressedFileNames) <= 3.0.4-1
  • rpmlib(FileDigests) <= 4.6.0-1
  • vulkan-headers
Provides1
  • clpeak = 2.1.4-1.fc45
ObsoletesNo Obsoletes
ConflictsNo Conflicts
RecommendsNo Recommends
SuggestsNo Suggests
SupplementsNo Supplements
EnhancesNo Enhances

Files

2
Files 1–2 of 2

Component of

buildroots that installed this RPM

No Buildroots