Use int64 as indices for set_kv_buffer (#3039)

This commit is contained in:
Lianmin Zheng
2025-01-21 19:46:09 -08:00
committed by GitHub
parent a42213dbd4
commit 3d8f1c9bcf
6 changed files with 30 additions and 37 deletions

View File

@@ -296,7 +296,7 @@ def fused_softcap_kernel(
n_elements,
BLOCK_SIZE: tl.constexpr,
):
pid = tl.program_id(0)
pid = tl.program_id(0).to(tl.int64)
block_start = pid * BLOCK_SIZE
offsets = block_start + tl.arange(0, BLOCK_SIZE)
mask = offsets < n_elements