|
Why the L2 cache hit rate in L1 store requests is 100%
|
|
0
|
191
|
April 18, 2025
|
|
How are pct_of_peak metrics calculated?
|
|
6
|
187
|
April 30, 2025
|
|
How to calculate achieved value in nsight compute's roofline for a kernel?
|
|
4
|
175
|
May 23, 2025
|
|
Memory Table Missing in ncu-ui only for H100
|
|
5
|
151
|
April 30, 2025
|
|
Error opening .ncu-rep files
|
|
2
|
168
|
May 26, 2025
|
|
NCU too slow and incomplete
|
|
4
|
169
|
May 1, 2025
|
|
Question about l1tex__data_pipe_lsu_wavefronts.avg
|
|
8
|
321
|
April 23, 2025
|
|
Timeout issue connecting ncu to a python application
|
|
2
|
281
|
April 24, 2025
|
|
Failed to initialize the profiler: LibraryNotLoaded. Check that a compatible driver library is loaded
|
|
2
|
148
|
March 30, 2025
|
|
Root user ncu error no permission to access GPU performance counters
|
|
7
|
354
|
March 31, 2025
|
|
Why is the sm__warps_active so high
|
|
3
|
171
|
April 21, 2025
|
|
Nsight Compute with MPI: ‘No Kernels Were Profiled’ Warning and Hanging Issue
|
|
3
|
150
|
March 31, 2025
|
|
The Roofline chart is not displaying in the NCU-UI after enabling the --set full flag
|
|
6
|
134
|
March 31, 2025
|
|
==ERROR== ERR_NVGPUCTRPERM - The user does not have permission to access NVIDIA GPU Performance Counters on the target device 0
|
|
3
|
205
|
April 23, 2025
|
|
Nsight compute
|
|
3
|
189
|
April 28, 2025
|
|
Nsight-compute failing remote deployment of files (arm64 mac host, x86-64 ubuntu target)
|
|
7
|
161
|
April 21, 2025
|
|
Qusetion about kernel warmup and replay control
|
|
2
|
150
|
April 28, 2025
|
|
Launching Conda environment + Python for Cuda profiling
|
|
3
|
141
|
April 21, 2025
|
|
Failed to deploy files
|
|
2
|
137
|
March 20, 2025
|
|
==ERROR== Unable to write to file matmul.ncu-rep. Please verify this file is not locked, and writable
|
|
2
|
126
|
April 21, 2025
|
|
Problem in collecting roofline plot
|
|
3
|
152
|
April 21, 2025
|
|
Nsight-compute (ncu-ui) failure on jetpack 6.2 jetson orin nano
|
|
15
|
236
|
April 26, 2025
|
|
The profiler returned an error code: 3221226505 (0xc0000409)
|
|
4
|
149
|
March 26, 2025
|
|
Maximum Tensor Core utilization
|
|
4
|
171
|
March 20, 2025
|
|
FP64 pipe active even with no double-precision types or instructions used
|
|
5
|
160
|
March 20, 2025
|
|
Dram__bytes_read.sum is !(n/a)
|
|
2
|
128
|
March 20, 2025
|
|
Issues about the time shown in ncu
|
|
4
|
133
|
March 19, 2025
|
|
Ncu no kernels profiled -- Target process xxx terminated before first instrumented API call
|
|
5
|
175
|
March 18, 2025
|
|
In case of using peer memory, How can I measure the L1 or L2 cache's value on operating GPU?
|
|
3
|
180
|
March 18, 2025
|
|
Problems with lts__t_requests_srcunit_tex_aperture_peer
|
|
7
|
748
|
March 18, 2025
|