简体繁体 English

每个网格的最大块数：CUDA

[英]Maximum blocks per grid:CUDA

原文 2011-05-18 17:52:01 2 2 cuda/ nvidia

What is the maximum number of blocks in a grid that can created per kernel launch? 每个内核启动可以创建的网格中的最大块数是多少？ I am slightly confused here since 从那以后我有点困惑

Now the compute capability table here says that there can be 65535 blocks per grid dimemsion in CUDA compute capability 2.0. 现在，这里的计算能力表表明，在CUDA计算能力2.0中，每个网格可以有65535个块。

Does that mean the total number of blocks = 65535*65535? 这是否意味着总块数= 65535 * 65535？

Or does it mean that you can rearrange at most 65535 into a 1d grid of 65536 blocks or 2d grid of sqrt(65535) * sqrt(65535) ? 或者它是否意味着您可以将最多65535重新排列为65536块的1d网格或sqrt（65535）* sqrt（65535）的2d网格？

Thank you. 谢谢。

2 个解决方案

65535 per dimension of the grid. 每个网格尺寸65535。 On compute 1.x cards, 1D and 2D grids are supported. 在计算1.x卡上，支持1D和2D网格。 On compute 2.x cards, 3D grids are also supported, so 65535, 65535 x 65535, and 65535 x 65535 x 65535 are the limits for Fermi (compute 2.x) cards. 在计算2.x卡上，还支持3D网格，因此65535,65535 x 65535和65535 x 65535 x 65535是Fermi（计算2.x）卡的限制。

我认为每格为65535 ..

GPU/CUDA：网格的最大块数和每个多处理器的最大驻留块数 - GPU/CUDA: Maximum number of blocks of a grid and Maximum number of resident blocks per multiprocessor

CUDA 确定每个块的线程数，每个网格的块数 - CUDA determining threads per block, blocks per grid

CUDA块的最大数量？ - Maximum number of CUDA blocks?

在CUDA中每个网格（x尺寸）发射超过65536个块 - launch more than 65536 blocks per grid (x dimension) in CUDA

CUDA：网格中的最大块数！= CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_X？ - CUDA: Maximum number of blocks in a grid != CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_X?

CUDA网格阻挡了Vs。 MPI进程的网格，每个进程只有一个块 - A grid of CUDA blocks Vs. a grid of MPI processes with a single block per process

CUDA估计每个块的线程和2D网格数据的块编号 - CUDA estimating threads per blocks and block numbers for 2D grid data

cuda 每个多处理器的占用最大活动块 == 0？ - cuda Occupancy Max Active Blocks Per Multiprocessor == 0?

CUDA 优化网格步幅循环的块数 - CUDA optimise number of blocks for grid stride loop

是否可以使用CUDA在网格中选择某些块？ - It is possible to select some blocks in grid with CUDA?

暂无

暂无

声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 GPU/CUDA：网格的最大块数和每个多处理器的最大驻留块数 - GPU/CUDA: Maximum number of blocks of a grid and Maximum number of resident blocks per multiprocessor CUDA 确定每个块的线程数，每个网格的块数 - CUDA determining threads per block, blocks per grid CUDA块的最大数量？ - Maximum number of CUDA blocks? 在CUDA中每个网格（x尺寸）发射超过65536个块 - launch more than 65536 blocks per grid (x dimension) in CUDA CUDA：网格中的最大块数！= CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_X？ - CUDA: Maximum number of blocks in a grid != CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_X? CUDA网格阻挡了Vs。 MPI进程的网格，每个进程只有一个块 - A grid of CUDA blocks Vs. a grid of MPI processes with a single block per process CUDA估计每个块的线程和2D网格数据的块编号 - CUDA estimating threads per blocks and block numbers for 2D grid data cuda 每个多处理器的占用最大活动块 == 0？ - cuda Occupancy Max Active Blocks Per Multiprocessor == 0? CUDA 优化网格步幅循环的块数 - CUDA optimise number of blocks for grid stride loop 是否可以使用CUDA在网格中选择某些块？ - It is possible to select some blocks in grid with CUDA?

相关标签

粤ICP备18138465号 © 2020-2024 STACKOOM.COM