xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-26 02:49:37 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] LoRA - Retire unused maxnreg LoRA kernel argument (#17677)

Browse Source
This commit is contained in:
Varun Sundar Rabindranath 2025-05-06 06:04:29 +05:30 committed by GitHub
parent 5941e0b7ea
commit 90bd2ae172
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 0 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/lora/ops/triton_ops/lora_expand_op.py
View File

@ -204,7 +204,6 @@ def _lora_expand(
NUM_WARPS = 4
NUM_CTAS = 1
NUM_STAGES = 2
MAX_NREG = None
EVEN_K = K % BLOCK_K == 0 # type: ignore
@ -258,7 +257,6 @@ def _lora_expand(
num_warps=NUM_WARPS,
num_ctas=NUM_CTAS,
num_stages=NUM_STAGES,
maxnreg=MAX_NREG,
)
return

2
vllm/lora/ops/triton_ops/lora_shrink_op.py
View File

@ -168,7 +168,6 @@ def _lora_shrink(
NUM_WARPS = 4
NUM_CTAS = 1
NUM_STAGES = 2
MAX_NREG = None
EVEN_K = K % (BLOCK_K * SPLIT_K) == 0 # type: ignore
@ -213,7 +212,6 @@ def _lora_shrink(
num_warps=NUM_WARPS,
num_ctas=NUM_CTAS,
num_stages=NUM_STAGES,
maxnreg=MAX_NREG,
)
return