StackOverflow Questions for Tag: nvprof

Saydon

Reputation: 27

CUDA more load transactions than store even though both are coalesced?

memorycudaprofilingtransposenvprof

Score: -2

Answers: 1

Antonello Cioffi

Reputation: 31

Very strange timing in Nvidia Visual profiler

c++cudatimingmulti-gpunvprof

Score: 1

Answers: 0

danny

Reputation: 1319

nvprof option for bandwidth

cudaprofilingnvprof

Score: 5

Answers: 1

dru10

Reputation: 43

Nsys CLI profiling guidance

cudaprofilingnsightnvprofnsight-systems

Score: 1

Answers: 2

Cherry Toska

Reputation: 181

Roofline Model with CUDA Manual vs. Nsight Compute

cudansightnvprofnsight-computeroofline

Score: 2

Answers: 1

n00b

Reputation: 177

Is it possible to see that kernel execution happened on Tensor Cores or not via nvprof (or some other method)?

cudagpunvidiacudnnnvprof

Score: 5

Answers: 2

Principio Tudisco

Reputation: 65

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

c++cudafortrannvprof

Score: 2

Answers: 1

travelingbones

Reputation: 8448

How to observe CUDA events and metrics for a subsection of an executable (e.g. only during a kernel execution time)?

cudaprofilingnvvpnvprof

Score: 9

Answers: 2

Filipe Aleixo

Reputation: 4242

nvprof is using all available GPU's when profiling python script

pythoncudaprofilingnvprof

Score: 0

Answers: 1

Ken Y-N

Reputation: 15018

Profiling arbitrary CUDA applications

linuxcudanvprofnvvp

Score: 4

Answers: 2

anroesti

Reputation: 11403

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

cudanvprof

Score: 3

Answers: 1

skm

Reputation: 5679

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

windowscudanvprofnvvp

Score: 1

Answers: 1

Fasil

Reputation: 1

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

cmatrixcudagpgpunvprof

Score: 0

Answers: 0

edhu

Reputation: 451

nv-nsight-cu-cli caused Tensorflow to fail

tensorflowgpunvidianvprofnsight-compute

Score: 0

Answers: 2

jakemdaly

Reputation: 87

What is redzone_checker? Profiling my tensorflow application on a GPU

tensorflowgpuprofilingnvprof

Score: 5

Answers: 2

김양곤

Reputation: 41

cuda kernel 'volta_sgemm_128x32_nn' means what?

multithreadinggridblocktorchnvprof

Score: 3

Answers: 0

Bojan Niceno

Reputation: 163

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

nvidiaopenaccnvprof

Score: 0

Answers: 1

Booo

Reputation: 503

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

cudagpuprofilernvprofnvvp

Score: 0

Answers: 1

ritter

Reputation: 7729

NVIDIA Visual Profiler: Insufficient kernel bounds data

cudanvprofnvvp

Score: 0

Answers: 1

nonsence90

Reputation: 21

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

c++cudanvidianvprof

Score: 2

Answers: 0

PreviousPage 1Next

StackOverflow Questions for Tag: nvprof

CUDA more load transactions than store even though both are coalesced?

Very strange timing in Nvidia Visual profiler

nvprof option for bandwidth

Nsys CLI profiling guidance

Roofline Model with CUDA Manual vs. Nsight Compute

Is it possible to see that kernel execution happened on Tensor Cores or not via nvprof (or some other method)?

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

How to observe CUDA events and metrics for a subsection of an executable (e.g. only during a kernel execution time)?

nvprof is using all available GPU&#39;s when profiling python script

Profiling arbitrary CUDA applications

What is a transaction and a request in the &#39;gld_transactions_per_request&#39; metric of the Cuda profiler?

Profilers (nvvp and nvprof) not showing &quot;Page Fault&quot; information

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

nv-nsight-cu-cli caused Tensorflow to fail

What is redzone_checker? Profiling my tensorflow application on a GPU

cuda kernel &#39;volta_sgemm_128x32_nn&#39; means what?

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

Meaning of the &quot;flop_count_sp&quot; and &quot;inst_fp_32&quot; metric in CUDA Profiler

NVIDIA Visual Profiler: Insufficient kernel bounds data

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

nvprof is using all available GPU's when profiling python script

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

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

cuda kernel 'volta_sgemm_128x32_nn' means what?

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

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