xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 10:46:08 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Support triton==3.3.0+git95326d9f for RTX 5090 (Unsloth + vLLM compatibility) (#15471)

Browse Source 
Co-authored-by: ServerAI <ai@exc-mad-ai.com>
This commit is contained in:
Antonio Gómez 2025-03-25 18:59:25 +01:00 committed by GitHub
parent 0a049c7d86
commit 5d8e1c9279
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/lora/ops/triton_ops/kernel_utils.py
View File

@ -130,7 +130,7 @@ def do_expand_kernel(
# Identify A and B block pointers
offset_k = tl.arange(0, BLOCK_K)
a_ptr = (cur_input_ptr + ram[:, None] * input_d1_stride +
offset_k[None, :] * input_d2_stride, )
offset_k[None, :] * input_d2_stride)
b_ptr = (cur_lora_ptr + cur_lora_d0_stride * lora_index +
offset_k[:, None] * cur_lora_d2_stride +
rbn[None, :] * cur_lora_d1_stride)