Script 'mail_helper' called by obssrc Hello community, here is the log from the commit of package clpeak for openSUSE:Factory checked in at 2026-05-17 18:57:03 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Comparing /work/SRC/openSUSE:Factory/clpeak (Old) and /work/SRC/openSUSE:Factory/.clpeak.new.1966 (New) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "clpeak" Sun May 17 18:57:03 2026 rev:8 rq:1353583 version:2.0.6 Changes: -------- --- /work/SRC/openSUSE:Factory/clpeak/clpeak.changes 2026-03-30 18:38:10.004282574 +0200 +++ /work/SRC/openSUSE:Factory/.clpeak.new.1966/clpeak.changes 2026-05-17 18:57:23.961709422 +0200 @@ -1,0 +2,91 @@ +Wed May 13 16:55:51 UTC 2026 - Martin Hauke <[email protected]> + +- Update to version 2.0.6: + * CUDA integer4/int8 sparse tests, cleanup. + * Runtime gate for opencl int8 dp test. + +------------------------------------------------------------------- +Mon May 11 09:37:24 UTC 2026 - Martin Hauke <[email protected]> + +- Update to version 2.0.5: + * [vulkan] reduce work size when CU is not known. + * Zero Vulkan atomic buffers before dispatch. + * Calibrate with single-iteration probes. + * [vulkan] remove local bandwidth v8. + * Increase max-time default to 500 ms. + * Initial commit for time based runs. + * Atomic throughput: bump ATOMIC_REPS 64 -> 256 (M1 was + under-measuring). + * Atomic throughput: cut ATOMIC_REPS 512 -> 64 to avoid GPU + watchdog. + * Vulkan: reduce fp64 dispatch size to avoid GPU watchdog on + RDNA4 + RADV + * Kernel latency: report submit failures instead of bogus + roundtrip times. + +------------------------------------------------------------------- +Fri May 08 18:14:51 UTC 2026 - Martin Hauke <[email protected]> + +- Update to version 2.0.3: + * Vulkan: fix inf/bogus bandwidth on Adreno (Turnip) + drivers + with broken timers. + * Atomic throughput: cut per-test work and unify metric names + across backends. + * Vulkan: split coopmat dispatch into FP and INT phases. + * Vulkan: expand compute test variants and add new tests from + other backends. + +------------------------------------------------------------------- +Thu May 07 14:13:12 UTC 2026 - Martin Hauke <[email protected]> + +- Update to version 2.0.2: + * Remove debug signing config from build file. + * Change ZERO_COPY_MULTIPLIER from 5.0 to 3.0. + * [cmake] version - don't use long version on release. + +------------------------------------------------------------------- +Thu May 07 08:28:25 UTC 2026 - Martin Hauke <[email protected]> + +- Update to version 2.0.0: + New Backends + * Vulkan Compute, bandwidth, atomics, cooperative matrix (tensor + core equivalent). + * CUDA Compute, bandwidth, WMMA/BMMA tensor cores, cuBLASLt + GEMM. + * OpenCL Unchanged baseline; now explicitly selectable via + --opencl + New Benchmarks + * Floating-point Compute + + BFloat16 (BF16) throughput. + + CUDA WMMA tensor cores: FP16, BF16, FP8-E4M3, FP8-E5M2, TF32, FP64 + + CUDA BMMA (binary matrix). + + Vulkan cooperative matrix: FP16, BF16, INT8, FP8 variants. + + Metal simdgroup matrix: FP16, BF16. + * Integer Compute + + INT8 dot-product (DP4a). + + Packed INT4 (emulated). + + CUDA WMMA INT8 / INT4. + * GEMM — vendor-SDK peak + + CUDA: cuBLASLt (FP32, TF32, FP16, BF16, FP8-E4M3, FP8-E5M2, + INT8, INT4). + + Metal: MPS / MPSGraph (FP32, FP16, BF16). + Memory & Latency + * Local (shared) memory bandwidth. + * Image / texture memory bandwidth. + * Host↔Device transfer bandwidth (Vulkan, CUDA). + * Atomic throughput. + CLI Changes + * Several flags from v1.x have been renamed or removed. + Output Format Changes + * XML — breaking schema change (format_version="2") + v1.x XML files are not accepted as --compare baselines. + Bug Fixes & Improvements + * Unified measurement methodology: batch timing replaces + per-iteration event timing across all backends. + * Dispatch sizes auto-scaled to GPU compute unit / SM count. + * RAII scope guard replaces manual resultScopeBegin() / + resultScopeEnd() calls. + * Baseline comparison via --compare FILE accepts any v2 JSON, + CSV, or XML file. + +------------------------------------------------------------------- Old: ---- clpeak-1.1.7.tar.xz New: ---- clpeak-2.0.6.tar.xz ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Other differences: ------------------ ++++++ clpeak.spec ++++++ --- /var/tmp/diff_new_pack.q4JepA/_old 2026-05-17 18:57:24.505731618 +0200 +++ /var/tmp/diff_new_pack.q4JepA/_new 2026-05-17 18:57:24.509731781 +0200 @@ -18,10 +18,10 @@ Name: clpeak -Version: 1.1.7 +Version: 2.0.6 Release: 0 Summary: Find peak OpenCL capacities like bandwidth & compute -License: SUSE-Public-Domain +License: LicenseRef-SUSE-Public-Domain Group: System/Benchmark URL: https://github.com/krrishnarraj/clpeak Source: %{name}-%{version}.tar.xz @@ -41,8 +41,7 @@ %autosetup %build -%cmake \ - -Wno-dev +%cmake %cmake_build %install ++++++ _service ++++++ --- /var/tmp/diff_new_pack.q4JepA/_old 2026-05-17 18:57:24.541733086 +0200 +++ /var/tmp/diff_new_pack.q4JepA/_new 2026-05-17 18:57:24.545733249 +0200 @@ -1,7 +1,7 @@ <services> <service name="tar_scm" mode="manual"> <param name="url">https://github.com/krrishnarraj/clpeak.git</param> - <param name="revision">1.1.7</param> + <param name="revision">2.0.6</param> <param name="scm">git</param> <param name="changesgenerate">enable</param> <param name="filename">clpeak</param> @@ -14,4 +14,3 @@ <service name="set_version" mode="manual"/> </services> - ++++++ _servicedata ++++++ --- /var/tmp/diff_new_pack.q4JepA/_old 2026-05-17 18:57:24.573734392 +0200 +++ /var/tmp/diff_new_pack.q4JepA/_new 2026-05-17 18:57:24.577734555 +0200 @@ -1,6 +1,6 @@ <servicedata> <service name="tar_scm"> <param name="url">https://github.com/krrishnarraj/clpeak.git</param> - <param name="changesrevision">3aa260ffe0df7391d113b16109df4f48aa3efd92</param></service></servicedata> + <param name="changesrevision">7e7a8ac4dae8bc5c6e384b79a47e985150285f72</param></service></servicedata> (No newline at EOF) ++++++ clpeak-1.1.7.tar.xz -> clpeak-2.0.6.tar.xz ++++++ ++++ 28412 lines of diff (skipped)
