cuda.tile.bid

cuda.tile.bid(axis)

Gets the index of current block.

Parameters:

axis (const int) – The axis of the block index space. Possible values are 0, 1, 2.

Return type:

int32

Examples

SnippetComplete Example

bidx = ct.bid(0)
bidy = ct.bid(1)
bidz = ct.bid(2)
print(f"Hello from block ({bidx}, {bidy}, {bidz})")

Copy

import cuda.tile as ct
import torch

@ct.kernel
def kernel():
    bidx = ct.bid(0)
    bidy = ct.bid(1)
    bidz = ct.bid(2)
    print(f"Hello from block ({bidx}, {bidy}, {bidz})")


torch.cuda.init()
ct.launch(torch.cuda.current_stream(), (1,), kernel, ())
torch.cuda.synchronize()

Output

Hello from block (0, 0, 0)