trove
trove copied to clipboard
bryancatanzaro
Full-speed Array of Structures access
on CUDA11 w/ MSVC 2019, I get a lot of warnings when using trove: ``` [build] F:\fem_performance_tests\extern\trove\trove/warp.h(42): warning: integer conversion resulted in a change of sign [build] [build] F:\fem_performance_tests\extern\trove\trove/shfl.h(53): warning:...
I noticed that the warp_id calculation within much of the code assumes a 1D thread block. Is this an assumed limitation when using trove? ```c++ int warp_id = threadIdx.x &...
can you provide some detailed samples for actual use? Thanks.
Hi everyone, i am trying to use array of structure in my project in cuda and to increase the performance i would like to use the trove-library. Unfortunately i could...
I am pretty sure that only load time should be considered for benchmarking SoA loads with/without shuffle involving. Please take a look on the suggested solution. I am pretty sure...
The following code crashes nvcc5. Compile with nvcc main.cu -Itrove-master -arch=compute_35 # include struct WU32 { unsigned int data[32]; }; __global__ void kernel(WU32\* in, WU32\* out) { int tid =...
