xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-24 03:37:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Misc] [ROCm] Prevent surplus tensor reshape (#19803)

Browse Source 
Signed-off-by: Zsolt Borbely <zsolt.borbely@htecgroup.com>
This commit is contained in:
zsolt-borbely-htec 2025-06-19 07:57:16 +02:00 committed by GitHub
parent 2de12be428
commit aa20d10a91
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/v1/attention/backends/triton_attn.py
View File

@ -376,7 +376,7 @@ class TritonAttentionImpl(AttentionImpl):
query.reshape( query.reshape(
(num_tokens, num_heads * head_size)).contiguous(), (num_tokens, num_heads * head_size)).contiguous(),
layer._q_scale) layer._q_scale)
query = query.reshape((num_tokens, num_heads, head_size)) query = query.reshape((num_tokens, num_heads, head_size))
use_local_attn = \ use_local_attn = \
(self.use_irope and attn_metadata.local_attn_metadata is not None) (self.use_irope and attn_metadata.local_attn_metadata is not None)