fix rms_norm #945

SangChengC · 2025-06-26T09:09:30Z

No description provided.

gemini-code-assist

Summary of Changes

Hello @SangChengC, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refactors the RMS normalization usage within the Llama model's inference components. It standardizes the RMSNorm operation by migrating to a shared, potentially optimized, rms_norm kernel and adapts the function calls to its updated signature, ensuring consistency and correctness in the normalization process.

Highlights

RMSNorm Implementation Update: The RMS normalization function used across the Llama model's inference layers has been updated. The previous rmsnorm_forward function, specific to Llama, has been replaced with a more general rms_norm function, now imported from lightllm.models.vit.triton_kernel.rms_norm_vit. This suggests a consolidation or improvement of the underlying kernel.
RMSNorm Function Signature Change: The updated rms_norm function now requires the use_custom_tensor_mananger=True argument. Additionally, in transformer_layer_infer.py, the out argument, which previously allowed specifying an explicit output tensor, has been removed from the function calls to rms_norm in both _att_norm and _ffn_norm.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

The pull request replaces the rmsnorm_forward function with rms_norm from the vit model in the llama model. The import paths and the hardcoded use_custom_tensor_mananger=True should be double checked.

gemini-code-assist · 2025-06-26T09:11:04Z

lightllm/models/llama/layer_infer/post_layer_infer.py

@@ -8,7 +8,7 @@
 from lightllm.models.llama.layer_weights.pre_and_post_layer_weight import LlamaPreAndPostLayerWeight
 from einops import rearrange
 from lightllm.models.llama.infer_struct import LlamaInferStateInfo
-from lightllm.models.llama.triton_kernel.rmsnorm import rmsnorm_forward
+from lightllm.models.vit.triton_kernel.rms_norm_vit import rms_norm


The import path seems incorrect. This change imports rms_norm from the vit model, but the file name indicates it should be for llama. Double check this import.

Suggested change

from lightllm.models.vit.triton_kernel.rms_norm_vit import rms_norm

from lightllm.models.llama.triton_kernel.rmsnorm import rmsnorm_forward

gemini-code-assist · 2025-06-26T09:11:04Z

lightllm/models/llama/layer_infer/transformer_layer_infer.py

@@ -14,7 +14,7 @@
 from lightllm.models.llama.triton_kernel.token_attention_nopad_att1 import token_att_fwd, token_att_fwd_int8k
 from lightllm.models.llama.triton_kernel.token_attention_nopad_softmax import token_softmax_fwd
 from lightllm.models.llama.triton_kernel.token_attention_nopad_reduceV import token_att_fwd2, token_att_fwd2_int8v
-from lightllm.models.llama.triton_kernel.rmsnorm import rmsnorm_forward
+from lightllm.models.vit.triton_kernel.rms_norm_vit import rms_norm


The import path seems incorrect. This change imports rms_norm from the vit model, but the file name indicates it should be for llama. Double check this import.

Suggested change

from lightllm.models.vit.triton_kernel.rms_norm_vit import rms_norm

from lightllm.models.llama.triton_kernel.rmsnorm import rmsnorm_forward

gemini-code-assist · 2025-06-26T09:11:04Z

lightllm/models/llama/layer_infer/post_layer_infer.py

@@ -25,7 +25,7 @@ def __init__(self, network_config, mode):
        return

    def _norm(self, input, infer_state, layer_weight: LlamaPreAndPostLayerWeight) -> torch.Tensor:
-        return rmsnorm_forward(input, layer_weight.final_norm_weight_, eps=self.eps_)
+        return rms_norm(input, layer_weight.final_norm_weight_, eps=self.eps_, use_custom_tensor_mananger=True)


Consider whether use_custom_tensor_mananger=True should be a configurable parameter or determined based on the execution environment, rather than being hardcoded.

return rms_norm(input, layer_weight.final_norm_weight_, eps=self.eps_)

gemini-code-assist · 2025-06-26T09:11:04Z

lightllm/models/llama/layer_infer/transformer_layer_infer.py

@@ -135,14 +135,14 @@ def _att_norm(
        self, input, infer_state: LlamaInferStateInfo, layer_weight: LlamaTransformerLayerWeight
    ) -> torch.Tensor:
        out = self.alloc_tensor(input.shape, input.dtype)
-        rmsnorm_forward(input, weight=layer_weight.att_norm_weight_.weight, eps=self.eps_, out=out)
+        rms_norm(input, weight=layer_weight.att_norm_weight_.weight, eps=self.eps_, use_custom_tensor_mananger=True)


Consider whether use_custom_tensor_mananger=True should be a configurable parameter or determined based on the execution environment, rather than being hardcoded.

rms_norm(input, weight=layer_weight.att_norm_weight_.weight, eps=self.eps_)

gemini-code-assist · 2025-06-26T09:11:05Z

lightllm/models/llama/layer_infer/transformer_layer_infer.py

        return out

    def _ffn_norm(
        self, input, infer_state: LlamaInferStateInfo, layer_weight: LlamaTransformerLayerWeight
    ) -> torch.Tensor:
        out = self.alloc_tensor(input.shape, input.dtype)
-        rmsnorm_forward(input, weight=layer_weight.ffn_norm_weight_.weight, eps=self.eps_, out=out)
+        rms_norm(input, weight=layer_weight.ffn_norm_weight_.weight, eps=self.eps_, use_custom_tensor_mananger=True)


Consider whether use_custom_tensor_mananger=True should be a configurable parameter or determined based on the execution environment, rather than being hardcoded.

rms_norm(input, weight=layer_weight.ffn_norm_weight_.weight, eps=self.eps_)

fix rms_norm

feb505b

gemini-code-assist bot reviewed Jun 26, 2025

View reviewed changes

SangChengC force-pushed the fix_rms_norm branch from 268c18e to 6655d1d Compare June 27, 2025 06:25

fix rms_norm

dc4475a

SangChengC force-pushed the fix_rms_norm branch 3 times, most recently from f432e97 to 45e2878 Compare June 27, 2025 08:25

fix rmsnorm

60efbef

SangChengC force-pushed the fix_rms_norm branch from 45e2878 to 60efbef Compare June 27, 2025 08:26

internvl-llm support qwen3_moe

689c431

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix rms_norm #945

fix rms_norm #945

Uh oh!

SangChengC commented Jun 26, 2025

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

gemini-code-assist bot Jun 26, 2025

Uh oh!

gemini-code-assist bot Jun 26, 2025

Uh oh!

gemini-code-assist bot Jun 26, 2025

Uh oh!

gemini-code-assist bot Jun 26, 2025

Uh oh!

gemini-code-assist bot Jun 26, 2025

Uh oh!

Uh oh!

	from lightllm.models.vit.triton_kernel.rms_norm_vit import rms_norm
	from lightllm.models.llama.triton_kernel.rmsnorm import rmsnorm_forward

fix rms_norm #945

Are you sure you want to change the base?

fix rms_norm #945

Uh oh!

Conversation

SangChengC commented Jun 26, 2025

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist bot Jun 26, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Jun 26, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Jun 26, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Jun 26, 2025

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist bot Jun 26, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!