• Julien's avatar
    Merge pull request #16218 from JulienMaille:cuda-dnn-for-older-gpus · 4e2ef8c8
    Julien authored
    Enable cuda4dnn on hardware without support for __half
    
    * Enable cuda4dnn on hardware without support for half (ie. compute capability < 5.3)
    
    Update CMakeLists.txt
    
    Lowered minimum CC to 3.0
    
    * UPD: added ifdef on new copy kernel
    
    * added fp16 support detection at runtime
    
    * Clarified #if condition on atomicAdd definition
    
    * More explicit CMake error message
    4e2ef8c8
Name
Last commit
Last update
..
activations.cu Loading commit data...
array.hpp Loading commit data...
atomics.hpp Loading commit data...
bias_activation.cu Loading commit data...
concat.cu Loading commit data...
crop_and_resize.cu Loading commit data...
eltwise_ops.cu Loading commit data...
execution.hpp Loading commit data...
fill_copy.cu Loading commit data...
fp_conversion.cu Loading commit data...
grid_stride_range.hpp Loading commit data...
kernel_dispatcher.hpp Loading commit data...
limits.hpp Loading commit data...
math.hpp Loading commit data...
max_unpooling.cu Loading commit data...
normalize.cu Loading commit data...
padding.cu Loading commit data...
permute.cu Loading commit data...
prior_box.cu Loading commit data...
region.cu Loading commit data...
resize.cu Loading commit data...
roi_pooling.cu Loading commit data...
scale_shift.cu Loading commit data...
slice.cu Loading commit data...
types.hpp Loading commit data...
vector_traits.hpp Loading commit data...