xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-15 11:36:20 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Hybrid] [Kernel] Fix chunk scan kernel when BLOCK_SIZE_DSTATE > 128 (#28295)

Browse Source 
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
This commit is contained in:
Thomas Parnell 2025-11-14 23:55:42 +01:00 committed by GitHub
parent bf3ffb61e6
commit e0c910bb89
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/mamba/ops/ssd_chunk_scan.py
View File

@ -245,7 +245,7 @@ def _chunk_scan_fwd_kernel(
) )
if not HAS_INITSTATES and (seq_idx != seq_idx_prev): if not HAS_INITSTATES and (seq_idx != seq_idx_prev):
prev_states = tl.zeros( prev_states = tl.zeros(
(BLOCK_SIZE_DSTATE, BLOCK_SIZE_K), dtype=C_ptr.dtype.element_ty (BLOCK_SIZE_K, BLOCK_SIZE_N), dtype=C_ptr.dtype.element_ty
) )
else: else:
prev_states = tl.load( prev_states = tl.load(