xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-23 09:24:28 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bug] Fix torch warning of tf32 usage (#29112)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye 2025-11-20 20:54:59 -05:00 committed by GitHub
parent ed6ae1e36a
commit e1eefa4c40
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/model_executor/layers/batch_invariant.py
View File

@ -852,5 +852,6 @@ def init_batch_invariance():
enable_batch_invariant_mode()
# Disable TF32 for batch invariance - it causes non-deterministic rounding
torch.backends.cuda.matmul.allow_tf32 = False
torch.backends.cudnn.allow_tf32 = False
torch.backends.cuda.matmul.fp32_precision = "ieee"
torch.backends.cudnn.conv.fp32_precision = "ieee"
torch.backends.cudnn.rnn.fp32_precision = "ieee"