edisonchan

fix error for cl.hpp and char * compare.

fix error: comparison between pointer and integer ('char *' and 'char')

fix error: comparison between pointer and integer ('char *' and 'char')

 **Is your feature request related to a problem? Please describe.** The test results currently being recorded are rounded to one decimal place. **Describe the solution you'd like** It would...

Package: 2047.5W when usleep(10000000); ?

1

just change usleep(100000); to usleep(10000000); sudo ./power_amd 0 (0), 1 (0), 2 (0), 3 (0), 4 (0), 5 (0), 6 (0), 7 (0) 8 (0), 9 (0), 10 (0), 11...

I had try to modify the asm code to aarch64, but not works. ``` .global data_test .type data_test, %function .data buffer: .skip 4096 page_cnt: .xword 0 total_ops: .xword 0 .text...

``` if (_strnicmp(argv[argIdx], "scalarlatency", 13) == 0) { ``` It would cause "I'm so confused. Unknown test type vectorlatency" if -test vectorlatency. it should be: ``` else if (_strnicmp(argv[argIdx], "scalarlatency",...

add support for Blackwell new NVDEC?

1

https://docs.nvidia.com/video-technologies/video-codec-sdk/13.0/read-me/index.html https://git.ffmpeg.org/gitweb/ffmpeg.git/commit/30e6effff94c6f4310aa2db571917bb2952f4d9e https://developer.nvidia.com/video-encode-and-decode-gpu-support-matrix-new The NVDEC of Blackwell add 4:2:2 support.

[QST] error: non-type template parameters of class type only available with ‘-std=c++20’ or ‘-std=gnu++20’

4

``` nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2025 NVIDIA Corporation Built on Fri_Feb_21_20:23:50_PST_2025 Cuda compilation tools, release 12.8, V12.8.93 Build cuda_12.8.r12.8/compiler.35583870_0 uname -a Linux u24 6.14.0-061400-generic...

The GPU clock is 8xxMHz only when running with RTX 5090 GPU. vsync was disabled in nvidia driver cp. The command line is: ``` RayTracer.exe --benchmark --width 3840 --height 2160...