Skip to content

Fix: rmsnorm_kernel config ignored when flash attention is not enabled #886

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
helloyongyang merged 1 commit into from
Feb 5, 2026
Merged

Fix: rmsnorm_kernel config ignored when flash attention is not enabled #886

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 10 additions & 10 deletions lightx2v/models/input_encoders/lightllm/qwen25_text_encoder_kernel.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -124,16 +124,16 @@ def _apply_kernel_optimizations(self):
if self.use_flash_attention_kernel:
logger.info(" ✓ Flash Attention 2 (loaded with model)")

if self.use_rmsnorm_kernel:
try:
from sgl_kernel.elementwise import rmsnorm

self._rmsnorm_kernel = rmsnorm
self._replace_rmsnorm_with_kernel()
logger.info(" ✓ RMSNorm kernel integrated (from sgl_kernel)")
except ImportError as e:
logger.warning(f" ✗ Failed to import sgl_kernel: {e}. RMSNorm optimization disabled.")
self.use_rmsnorm_kernel = False
if self.use_rmsnorm_kernel:
try:
from sgl_kernel.elementwise import rmsnorm

self._rmsnorm_kernel = rmsnorm
self._replace_rmsnorm_with_kernel()
logger.info(" ✓ RMSNorm kernel integrated (from sgl_kernel)")
except ImportError as e:
logger.warning(f" ✗ Failed to import sgl_kernel: {e}. RMSNorm optimization disabled.")
self.use_rmsnorm_kernel = False

def _replace_rmsnorm_with_kernel(self):
"""Replace RMSNorm layers with fused kernel"""
Expand Down