|
Unusual Latency Discrepancy between NCU and Standalone %clock64 (SM 8.9)
|
|
5
|
87
|
April 2, 2026
|
|
Ncu not running on WSL2 Ubuntu
|
|
2
|
73
|
May 26, 2026
|
|
Ncu unable to deploy stock section files to default directory because the directory is not writable
|
|
5
|
72
|
April 15, 2026
|
|
Discrepancy between throughput and roofline chart?
|
|
4
|
99
|
October 21, 2025
|
|
Hit rate is not as expected for a simple kernel
|
|
3
|
94
|
October 29, 2025
|
|
Two output scoreboard dependencies for instruction involving only one register
|
|
0
|
105
|
October 2, 2025
|
|
Nsight Compute runs into undefined symbol error or Slurm PMI error when profiling NVSHMEM
|
|
6
|
131
|
August 19, 2025
|
|
Matrix transpose perfomance profile explanation
|
|
9
|
317
|
April 26, 2025
|
|
A lot of stalls even with 100% occupancy
|
|
11
|
258
|
April 22, 2025
|
|
How to use Nsight Compute to profile a single CUDA kernel on different processes
|
|
4
|
192
|
April 10, 2025
|