NVBIO
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
Classes | Namespaces | Functions
arch.h File Reference
#include <nvbio/basic/types.h>
#include <nvbio/basic/numbers.h>
#include <nvbio/basic/console.h>
#include <nvbio/basic/exceptions.h>
#include <cuda_runtime.h>
#include <thrust/version.h>
#include <thrust/device_vector.h>
#include <thrust/host_vector.h>
#include <nvbio/basic/cuda/arch_inl.h>

Go to the source code of this file.

Classes

struct  nvbio::cuda::Arch
 

Namespaces

 nvbio
 Define a vector_view POD type and plain_view() for std::vector.
 
 nvbio::cuda
 

Functions

void nvbio::cuda::device_arch (uint32 &major, uint32 &minor)
 
uint32 nvbio::cuda::max_grid_size ()
 
size_t nvbio::cuda::multiprocessor_count ()
 
size_t nvbio::cuda::smem_allocation_unit (const cudaDeviceProp &properties)
 
size_t nvbio::cuda::reg_allocation_unit (const cudaDeviceProp &properties, const size_t regsPerThread)
 
size_t nvbio::cuda::warp_allocation_multiple (const cudaDeviceProp &properties)
 
size_t nvbio::cuda::num_sides_per_multiprocessor (const cudaDeviceProp &properties)
 
size_t nvbio::cuda::max_blocks_per_multiprocessor (const cudaDeviceProp &properties)
 
size_t nvbio::cuda::num_regs_per_block (const cudaDeviceProp &properties, const cudaFuncAttributes &attributes, const size_t CTA_SIZE)
 
template<typename KernelFunction >
cudaFuncAttributes nvbio::cuda::function_attributes (KernelFunction kernel)
 
template<typename KernelFunction >
size_t nvbio::cuda::max_active_blocks_per_multiprocessor (KernelFunction kernel, const size_t CTA_SIZE, const size_t dynamic_smem_bytes)
 
template<typename KernelFunction >
size_t nvbio::cuda::max_active_blocks (KernelFunction kernel, const size_t CTA_SIZE, const size_t dynamic_smem_bytes)
 
template<typename KernelFunction >
size_t nvbio::cuda::num_registers (KernelFunction kernel)
 
template<typename KernelFunction >
size_t nvbio::cuda::max_blocksize_with_highest_occupancy (KernelFunction kernel, size_t dynamic_smem_bytes_per_thread)
 
bool nvbio::cuda::is_tcc_enabled ()
 
void nvbio::cuda::check_error (const char *message)
 
template<uint32 N>
NVBIO_FORCEINLINE
NVBIO_HOST_DEVICE void 
nvbio::cuda::syncthreads ()