Question

Is there a way to enable `-Wconversion` in nvcc for the device code (the kernel code)?

score 42 · Answer 1 · 2024-03-17T19:53:20.117000

42

Views

Is there a way to enable `-Wconversion` in nvcc for the device code (the kernel code)?

Published on 17 March 2024 at 19:53

score 35 · Answer 2 · 2024-03-11T13:10:41.257000

35

Views

Compilation Error with Boost.Geometry and NVCC (CUDA 12.4): '__T0' does not name a type

Published on 11 March 2024 at 13:10

score 48 · Answer 3 · 2024-03-08T18:38:59.323000

48

Views

Compute and Data transfer not happening concurrently in cuda Streams on Iteration 2

Published on 08 March 2024 at 18:38

score 45 · Answer 4 · 2024-02-27T21:50:39.433000

45

Views

Missing functions for NVIDIA CUDA 12.3 CSR Formatted Sparse Linear Algebra

Published on 27 February 2024 at 21:50

score 38 · Answer 5 · 2024-02-27T16:31:18.480000

38

Views

On windows Server 2019 nvidia-smi cannot show the correct installed version of CUDA

Published on 27 February 2024 at 16:31

score 95 · Answer 6 · 2024-02-20T11:45:26.367000

95

Views

Error linking c++ shared library with CUDA

Published on 20 February 2024 at 11:45

score 76 · Answer 7 · 2024-02-18T23:37:15.237000

76

Views

CMake fails to link a shared library decorated with a suffix (like somelib.so.1). Is there any way around?

Published on 18 February 2024 at 23:37

score 63 · Answer 8 · 2024-02-17T22:12:56.427000

63

Views

Compiling cpp helper function with main.cu using NVCC compiler

Published on 17 February 2024 at 22:12

score 87 · Answer 9 · 2024-02-01T04:20:02.910000

87

Views

Register usage count of kernel different with and without -lineinfo flag

Published on 01 February 2024 at 04:20

score 54 · Answer 10 · 2024-01-22T21:50:17.077000

54

Views

CMake and Cuda separate compilation of class constructor on device fail during linking

Published on 22 January 2024 at 21:50

score 163 · Answer 11 · 2024-01-22T05:31:42.153000

163

Views

couldn't allocate an array of size 2116800 on device code

Published on 22 January 2024 at 05:31

score 40 · Answer 12 · 2024-01-08T16:21:51.557000

40

Views

L1/Texture cache enabling effect on nvcc(cuda 10.2) on jetson nano (maxwell architecture)

Published on 08 January 2024 at 16:21

score 172 · Answer 13 · 2024-01-08T13:21:10.637000

172

Views

clangd mangling system include paths

Published on 08 January 2024 at 13:21

score 81 · Answer 14 · 2024-01-02T16:27:14.090000

81

Views

How do I include Cutlass in a CuPy project?

Published on 02 January 2024 at 16:27

score 126 · Answer 15 · 2023-12-17T16:01:35.697000

126

Views

CondaValueError: invalid package specification: nvcc --version=10.1

Published on 17 December 2023 at 16:01

score 93 · Answer 16 · 2023-11-13T12:56:24.160000

93

Views

Compilation Errors with CUDA and MSVC: Identifiers from <atomic> Undefined

Published on 13 November 2023 at 12:56

score 28 · Answer 17 · 2023-11-09T03:16:46.890000

28

Views

icpx: error: cannot specify -o when generating multiple output files

Published on 09 November 2023 at 03:16

score 102 · Answer 18 · 2023-11-03T14:37:56.457000

102

Views

Why does this code not compile in CUDA c++20 standard?

Published on 03 November 2023 at 14:37

score 2.1k · Answer 19 · 2023-10-06T16:12:21.917000

2.1k

Views

Getting CUDA version correctly reported by nvcc

Published on 06 October 2023 at 16:12

score 292 · Answer 20 · 2023-09-25T07:55:13.500000

292

Views

How to compile C++20 CUDA program on Ubuntu 20?

Published on 25 September 2023 at 07:55

List Question

Is there a way to enable `-Wconversion` in nvcc for the device code (the kernel code)?

Compilation Error with Boost.Geometry and NVCC (CUDA 12.4): '__T0' does not name a type

Compute and Data transfer not happening concurrently in cuda Streams on Iteration 2

Missing functions for NVIDIA CUDA 12.3 CSR Formatted Sparse Linear Algebra

On windows Server 2019 nvidia-smi cannot show the correct installed version of CUDA

Error linking c++ shared library with CUDA

CMake fails to link a shared library decorated with a suffix (like somelib.so.1). Is there any way around?

Compiling cpp helper function with main.cu using NVCC compiler

Register usage count of kernel different with and without -lineinfo flag

CMake and Cuda separate compilation of class constructor on device fail during linking

couldn't allocate an array of size 2116800 on device code

L1/Texture cache enabling effect on nvcc(cuda 10.2) on jetson nano (maxwell architecture)

clangd mangling system include paths

How do I include Cutlass in a CuPy project?

CondaValueError: invalid package specification: nvcc --version=10.1

Compilation Errors with CUDA and MSVC: Identifiers from <atomic> Undefined

icpx: error: cannot specify -o when generating multiple output files

Why does this code not compile in CUDA c++20 standard?

Getting CUDA version correctly reported by nvcc

How to compile C++20 CUDA program on Ubuntu 20?

Trending Questions

Popular # Hahtags

Popular Questions