Question

Very strange timing in Nvidia Visual profiler

score 33 · Answer 1 · 2024-01-13T18:33:41.330000

33

Views

Very strange timing in Nvidia Visual profiler

Published on 13 January 2024 at 18:33

score 356 · Answer 2 · 2023-07-12T19:10:27.310000

356

Views

Roofline Model with CUDA Manual vs. Nsight Compute

Published on 12 July 2023 at 19:10

score 4.3k · Answer 3 · 2023-05-19T19:29:09.807000

4.3k

Views

Nsys CLI profiling guidance

Published on 19 May 2023 at 19:29

score 124 · Answer 4 · 2022-10-26T04:40:58.523000

124

Views

nvprof --metrics works with c++ executable but not with fortran executable

Published on 26 October 2022 at 04:40

score 303 · Answer 5 · 2021-11-29T12:12:01.607000

303

Views

Profilers (nvvp and nvprof) not showing "Page Fault" information

Published on 29 November 2021 at 12:12

score 39 · Answer 6 · 2021-10-19T16:35:44.383000

39

Views

Using Matrix addition in cuda c,code executes but when profiling it with nvprof.It says NO kernels are profiled

Published on 19 October 2021 at 16:35

score 401 · Answer 7 · 2021-02-23T06:16:04.553000

401

Views

cuda kernel 'volta_sgemm_128x32_nn' means what?

Published on 23 February 2021 at 06:16

score 1.2k · Answer 8 · 2020-09-24T10:30:32.113000

1.2k

Views

nvprof Warning: The path to CUPTI and CUDA Injection libraries might not be set in LD_LIBRARY_PATH

Published on 24 September 2020 at 10:30

score 334 · Answer 9 · 2020-09-09T17:06:58.820000

334

Views

Meaning of the "flop_count_sp" and "inst_fp_32" metric in CUDA Profiler

Published on 09 September 2020 at 17:06

score 587 · Answer 10 · 2020-08-18T22:08:34.027000

587

Views

NVIDIA Visual Profiler: Insufficient kernel bounds data

Published on 18 August 2020 at 22:08

score 882 · Answer 11 · 2020-08-18T07:54:28.847000

882

Views

CUDA nvprof on Windows: "Warning: unable to locate profiling library, GPU profiling skipped" (NOT cupti64_102.dll)

Published on 18 August 2020 at 07:54

score 157 · Answer 12 · 2020-07-30T06:20:42.273000

157

Views

Why don't I get "thread_inst_executed"

Published on 30 July 2020 at 06:20

score 145 · Answer 13 · 2020-07-14T03:45:28.680000

145

Views

dram_write_bytes result on P100

Published on 14 July 2020 at 03:45

score 4.3k · Answer 14 · 2020-05-18T08:16:05.090000

4.3k

Views

nvprof command error: cupti64_102.dll was not found

Published on 18 May 2020 at 08:16

score 1.6k · Answer 15 · 2020-03-27T07:09:57.543000

1.6k

Views

Running nvprof --metrics command under windows gives an error：cuda profiling error

Published on 27 March 2020 at 07:09

score 637 · Answer 16 · 2020-03-25T16:45:54.933000

637

Views

How to capture GPU data when profiling Tensorflow code with nvprof?

Published on 25 March 2020 at 16:45

score 863 · Answer 17 · 2020-03-24T05:32:44.713000

863

Views

What is redzone_checker? Profiling my tensorflow application on a GPU

Published on 24 March 2020 at 05:32

score 188 · Answer 18 · 2020-03-16T07:30:40.867000

188

Views

How to stop running TensorRT server without using ctrl-c (for profiling with nvprof)

Published on 16 March 2020 at 07:30

score 590 · Answer 19 · 2020-03-04T22:50:22.410000

590

Views

What is a transaction and a request in the 'gld_transactions_per_request' metric of the Cuda profiler?

Published on 04 March 2020 at 22:50

score 28 · Answer 20 · 2020-02-27T15:44:10.520000

28

Views

What is the reason for K80 versus Pascal performance differences in this program that adds two arrays?

Published on 27 February 2020 at 15:44

List Question

Very strange timing in Nvidia Visual profiler

Roofline Model with CUDA Manual vs. Nsight Compute

Nsys CLI profiling guidance

nvprof --metrics works with c++ executable but not with fortran executable

Profilers (nvvp and nvprof) not showing "Page Fault" information

Using Matrix addition in cuda c,code executes but when profiling it with nvprof.It says NO kernels are profiled

cuda kernel 'volta_sgemm_128x32_nn' means what?

nvprof Warning: The path to CUPTI and CUDA Injection libraries might not be set in LD_LIBRARY_PATH

Meaning of the "flop_count_sp" and "inst_fp_32" metric in CUDA Profiler

NVIDIA Visual Profiler: Insufficient kernel bounds data

CUDA nvprof on Windows: "Warning: unable to locate profiling library, GPU profiling skipped" (NOT cupti64_102.dll)

Why don't I get "thread_inst_executed"

dram_write_bytes result on P100

nvprof command error: cupti64_102.dll was not found

Running nvprof --metrics command under windows gives an error：cuda profiling error

How to capture GPU data when profiling Tensorflow code with nvprof?

What is redzone_checker? Profiling my tensorflow application on a GPU

How to stop running TensorRT server without using ctrl-c (for profiling with nvprof)

What is a transaction and a request in the 'gld_transactions_per_request' metric of the Cuda profiler?

What is the reason for K80 versus Pascal performance differences in this program that adds two arrays?

Trending Questions

Popular # Hahtags

Popular Questions