MCPcopy Create free account
hub / github.com/NVlabs/tiny-cuda-nn / GPUMatrixDynamic

Method GPUMatrixDynamic

include/tiny-cuda-nn/gpu_matrix.h:113–118  ·  view source on GitHub ↗

Owning its memory as a GPUMemory

Source from the content-addressed store, hash-verified

111
112 // Owning its memory as a GPUMemory<T>
113 GPUMatrixDynamic(uint32_t m, uint32_t n, MatrixLayout layout = CM)
114 : m_rows{m}, m_cols{n}, m_layout{layout} {
115 m_malloc_allocation = std::make_shared<GPUMemory<uint8_t>>(m * n * sizeof(T));
116 m_data = (T*)m_malloc_allocation->data();
117 set_stride_contiguous();
118 }
119
120 // Owning its memory as an allocation from a stream's memory arena
121 GPUMatrixDynamic(uint32_t m, uint32_t n, cudaStream_t stream, MatrixLayout layout = CM)

Callers

nothing calls this directly

Calls 2

allocate_workspaceFunction · 0.85
dataMethod · 0.45

Tested by

no test coverage detected